| { | |
| "best_metric": 1.631639838218689, | |
| "best_model_checkpoint": "flan-t5-small-samsum/checkpoint-3684", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 7368, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.728555917480999e-05, | |
| "loss": 1.5521, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.457111834961998e-05, | |
| "loss": 1.5546, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 4.185667752442997e-05, | |
| "loss": 1.5782, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 16.80952380952381, | |
| "eval_loss": 1.6427778005599976, | |
| "eval_rouge1": 43.7392, | |
| "eval_rouge2": 19.2055, | |
| "eval_rougeL": 36.2978, | |
| "eval_rougeLsum": 39.9317, | |
| "eval_runtime": 45.0957, | |
| "eval_samples_per_second": 18.161, | |
| "eval_steps_per_second": 2.284, | |
| "step": 1842 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 3.914223669923996e-05, | |
| "loss": 1.5701, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 3.642779587404995e-05, | |
| "loss": 1.5175, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 3.3713355048859935e-05, | |
| "loss": 1.5559, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 3.099891422366993e-05, | |
| "loss": 1.5494, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 16.936507936507937, | |
| "eval_loss": 1.631639838218689, | |
| "eval_rouge1": 44.0128, | |
| "eval_rouge2": 19.4478, | |
| "eval_rougeL": 36.4456, | |
| "eval_rougeLsum": 40.0738, | |
| "eval_runtime": 43.2682, | |
| "eval_samples_per_second": 18.928, | |
| "eval_steps_per_second": 2.381, | |
| "step": 3684 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 2.8284473398479917e-05, | |
| "loss": 1.5269, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 2.5570032573289905e-05, | |
| "loss": 1.5179, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 2.2855591748099893e-05, | |
| "loss": 1.5423, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 2.0141150922909884e-05, | |
| "loss": 1.5445, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 16.943833943833944, | |
| "eval_loss": 1.6355613470077515, | |
| "eval_rouge1": 44.0056, | |
| "eval_rouge2": 19.5202, | |
| "eval_rougeL": 36.3341, | |
| "eval_rougeLsum": 39.9956, | |
| "eval_runtime": 44.6582, | |
| "eval_samples_per_second": 18.339, | |
| "eval_steps_per_second": 2.306, | |
| "step": 5526 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 1.742671009771987e-05, | |
| "loss": 1.5163, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 1.471226927252986e-05, | |
| "loss": 1.5263, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 1.1997828447339848e-05, | |
| "loss": 1.5379, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 16.964590964590965, | |
| "eval_loss": 1.6338579654693604, | |
| "eval_rouge1": 43.8323, | |
| "eval_rouge2": 19.6412, | |
| "eval_rougeL": 36.2335, | |
| "eval_rougeLsum": 39.9361, | |
| "eval_runtime": 43.7905, | |
| "eval_samples_per_second": 18.703, | |
| "eval_steps_per_second": 2.352, | |
| "step": 7368 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 9210, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 1.0954155661197312e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |