| { |
| "best_metric": 1.6302473545074463, |
| "best_model_checkpoint": "./results/t5-bow-decoder-base\\checkpoint-6750", |
| "epoch": 3.0, |
| "global_step": 6750, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.11, |
| "learning_rate": 0.0005, |
| "loss": 3.0243, |
| "step": 256 |
| }, |
| { |
| "epoch": 0.23, |
| "learning_rate": 0.001, |
| "loss": 1.9017, |
| "step": 512 |
| }, |
| { |
| "epoch": 0.34, |
| "learning_rate": 0.0009942456392735119, |
| "loss": 1.8801, |
| "step": 768 |
| }, |
| { |
| "epoch": 0.46, |
| "learning_rate": 0.000988491278547024, |
| "loss": 1.875, |
| "step": 1024 |
| }, |
| { |
| "epoch": 0.57, |
| "learning_rate": 0.0009827369178205358, |
| "loss": 1.7517, |
| "step": 1280 |
| }, |
| { |
| "epoch": 0.68, |
| "learning_rate": 0.000976982557094048, |
| "loss": 1.8016, |
| "step": 1536 |
| }, |
| { |
| "epoch": 0.8, |
| "learning_rate": 0.0009712281963675598, |
| "loss": 1.7453, |
| "step": 1792 |
| }, |
| { |
| "epoch": 0.91, |
| "learning_rate": 0.0009654738356410718, |
| "loss": 1.7931, |
| "step": 2048 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 1.6831984519958496, |
| "eval_runtime": 18.9287, |
| "eval_samples_per_second": 52.83, |
| "eval_steps_per_second": 6.604, |
| "step": 2250 |
| }, |
| { |
| "epoch": 1.02, |
| "learning_rate": 0.0009597194749145838, |
| "loss": 1.7622, |
| "step": 2304 |
| }, |
| { |
| "epoch": 1.14, |
| "learning_rate": 0.0009539651141880957, |
| "loss": 1.5755, |
| "step": 2560 |
| }, |
| { |
| "epoch": 1.25, |
| "learning_rate": 0.0009482107534616077, |
| "loss": 1.6284, |
| "step": 2816 |
| }, |
| { |
| "epoch": 1.37, |
| "learning_rate": 0.0009424563927351196, |
| "loss": 1.6671, |
| "step": 3072 |
| }, |
| { |
| "epoch": 1.48, |
| "learning_rate": 0.0009367020320086316, |
| "loss": 1.5683, |
| "step": 3328 |
| }, |
| { |
| "epoch": 1.59, |
| "learning_rate": 0.0009309476712821435, |
| "loss": 1.618, |
| "step": 3584 |
| }, |
| { |
| "epoch": 1.71, |
| "learning_rate": 0.0009251933105556555, |
| "loss": 1.6381, |
| "step": 3840 |
| }, |
| { |
| "epoch": 1.82, |
| "learning_rate": 0.0009194389498291674, |
| "loss": 1.6486, |
| "step": 4096 |
| }, |
| { |
| "epoch": 1.93, |
| "learning_rate": 0.0009136845891026794, |
| "loss": 1.6367, |
| "step": 4352 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 1.6470595598220825, |
| "eval_runtime": 19.0654, |
| "eval_samples_per_second": 52.451, |
| "eval_steps_per_second": 6.556, |
| "step": 4500 |
| }, |
| { |
| "epoch": 2.05, |
| "learning_rate": 0.0009079302283761914, |
| "loss": 1.527, |
| "step": 4608 |
| }, |
| { |
| "epoch": 2.16, |
| "learning_rate": 0.0009021758676497032, |
| "loss": 1.4307, |
| "step": 4864 |
| }, |
| { |
| "epoch": 2.28, |
| "learning_rate": 0.0008964215069232152, |
| "loss": 1.4458, |
| "step": 5120 |
| }, |
| { |
| "epoch": 2.39, |
| "learning_rate": 0.0008906671461967272, |
| "loss": 1.4737, |
| "step": 5376 |
| }, |
| { |
| "epoch": 2.5, |
| "learning_rate": 0.0008849127854702393, |
| "loss": 1.4935, |
| "step": 5632 |
| }, |
| { |
| "epoch": 2.62, |
| "learning_rate": 0.0008791584247437511, |
| "loss": 1.4757, |
| "step": 5888 |
| }, |
| { |
| "epoch": 2.73, |
| "learning_rate": 0.0008734040640172631, |
| "loss": 1.4712, |
| "step": 6144 |
| }, |
| { |
| "epoch": 2.84, |
| "learning_rate": 0.0008676497032907751, |
| "loss": 1.4494, |
| "step": 6400 |
| }, |
| { |
| "epoch": 2.96, |
| "learning_rate": 0.000861895342564287, |
| "loss": 1.4335, |
| "step": 6656 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 1.6302473545074463, |
| "eval_runtime": 19.2598, |
| "eval_samples_per_second": 51.922, |
| "eval_steps_per_second": 6.49, |
| "step": 6750 |
| } |
| ], |
| "max_steps": 45000, |
| "num_train_epochs": 20, |
| "total_flos": 1.644186304512e+16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
