| { | |
| "best_metric": 0.42519670724868774, | |
| "best_model_checkpoint": "models/flan-t5-large-multiquestions-fixed-prompt/checkpoint-6538", | |
| "epoch": 2.0, | |
| "global_step": 6538, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0001, | |
| "loss": 7.6159, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 9.462769958096057e-05, | |
| "loss": 0.4167, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 8.925539916192114e-05, | |
| "loss": 0.4268, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 8.38830987428817e-05, | |
| "loss": 0.4209, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 7.851079832384227e-05, | |
| "loss": 0.426, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 7.313849790480284e-05, | |
| "loss": 0.408, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.4266486167907715, | |
| "eval_runtime": 16.4767, | |
| "eval_samples_per_second": 30.346, | |
| "eval_steps_per_second": 3.824, | |
| "step": 3269 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 6.776619748576341e-05, | |
| "loss": 0.369, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 6.239389706672397e-05, | |
| "loss": 0.3493, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 5.702159664768454e-05, | |
| "loss": 0.3531, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 5.164929622864511e-05, | |
| "loss": 0.3433, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 4.6276995809605676e-05, | |
| "loss": 0.3419, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 4.090469539056624e-05, | |
| "loss": 0.345, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 3.553239497152681e-05, | |
| "loss": 0.3503, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.42519670724868774, | |
| "eval_runtime": 13.4978, | |
| "eval_samples_per_second": 37.043, | |
| "eval_steps_per_second": 4.667, | |
| "step": 6538 | |
| } | |
| ], | |
| "max_steps": 9807, | |
| "num_train_epochs": 3, | |
| "total_flos": 2.3780091548860416e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |