| { |
| "best_global_step": 1044, |
| "best_metric": 37.0629, |
| "best_model_checkpoint": "./results_pegasus/checkpoint-1044", |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 1044, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.14398848092152627, |
| "grad_norm": 71.70013427734375, |
| "learning_rate": 2.859195402298851e-05, |
| "loss": 29.788212890625, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.28797696184305255, |
| "grad_norm": 73.66615295410156, |
| "learning_rate": 2.7155172413793102e-05, |
| "loss": 24.89407958984375, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.4319654427645788, |
| "grad_norm": 70.73384094238281, |
| "learning_rate": 2.57183908045977e-05, |
| "loss": 24.67899169921875, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.5759539236861051, |
| "grad_norm": 67.04731750488281, |
| "learning_rate": 2.4281609195402298e-05, |
| "loss": 23.75695068359375, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.7199424046076314, |
| "grad_norm": 62.45637130737305, |
| "learning_rate": 2.2844827586206897e-05, |
| "loss": 23.85019287109375, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.8639308855291576, |
| "grad_norm": 1750.8475341796875, |
| "learning_rate": 2.1408045977011497e-05, |
| "loss": 23.247626953125, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 2.6326208114624023, |
| "eval_model_preparation_time": 0.0127, |
| "eval_rouge1": 42.9633, |
| "eval_rouge2": 20.797, |
| "eval_rougeL": 36.68, |
| "eval_rougeLsum": 36.5897, |
| "eval_runtime": 183.6557, |
| "eval_samples_per_second": 1.889, |
| "eval_steps_per_second": 1.889, |
| "step": 348 |
| }, |
| { |
| "epoch": 1.005759539236861, |
| "grad_norm": 85.54132843017578, |
| "learning_rate": 1.997126436781609e-05, |
| "loss": 22.84858642578125, |
| "step": 350 |
| }, |
| { |
| "epoch": 1.1497480201583874, |
| "grad_norm": 70.81609344482422, |
| "learning_rate": 1.853448275862069e-05, |
| "loss": 22.126240234375, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.2937365010799136, |
| "grad_norm": 214.08627319335938, |
| "learning_rate": 1.709770114942529e-05, |
| "loss": 21.31904052734375, |
| "step": 450 |
| }, |
| { |
| "epoch": 1.43772498200144, |
| "grad_norm": 77.83561706542969, |
| "learning_rate": 1.5660919540229885e-05, |
| "loss": 22.437314453125, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.5817134629229663, |
| "grad_norm": 123.08596801757812, |
| "learning_rate": 1.4224137931034483e-05, |
| "loss": 21.7467578125, |
| "step": 550 |
| }, |
| { |
| "epoch": 1.7257019438444925, |
| "grad_norm": 60.972755432128906, |
| "learning_rate": 1.2787356321839081e-05, |
| "loss": 21.26820068359375, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.8696904247660187, |
| "grad_norm": 90.81558227539062, |
| "learning_rate": 1.1350574712643677e-05, |
| "loss": 21.498505859375, |
| "step": 650 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 2.586069345474243, |
| "eval_model_preparation_time": 0.0127, |
| "eval_rouge1": 43.4483, |
| "eval_rouge2": 21.3048, |
| "eval_rougeL": 36.9716, |
| "eval_rougeLsum": 36.8742, |
| "eval_runtime": 179.8863, |
| "eval_samples_per_second": 1.929, |
| "eval_steps_per_second": 1.929, |
| "step": 696 |
| }, |
| { |
| "epoch": 2.011519078473722, |
| "grad_norm": 79.58966064453125, |
| "learning_rate": 9.913793103448277e-06, |
| "loss": 21.545439453125, |
| "step": 700 |
| }, |
| { |
| "epoch": 2.1555075593952484, |
| "grad_norm": 62.96016311645508, |
| "learning_rate": 8.477011494252873e-06, |
| "loss": 21.18465576171875, |
| "step": 750 |
| }, |
| { |
| "epoch": 2.299496040316775, |
| "grad_norm": 114.29676818847656, |
| "learning_rate": 7.040229885057472e-06, |
| "loss": 20.94623291015625, |
| "step": 800 |
| }, |
| { |
| "epoch": 2.443484521238301, |
| "grad_norm": 427.0086364746094, |
| "learning_rate": 5.603448275862069e-06, |
| "loss": 21.252041015625, |
| "step": 850 |
| }, |
| { |
| "epoch": 2.587473002159827, |
| "grad_norm": 82.54395294189453, |
| "learning_rate": 4.166666666666667e-06, |
| "loss": 20.805263671875, |
| "step": 900 |
| }, |
| { |
| "epoch": 2.7314614830813535, |
| "grad_norm": 195.64202880859375, |
| "learning_rate": 2.7298850574712644e-06, |
| "loss": 20.5162548828125, |
| "step": 950 |
| }, |
| { |
| "epoch": 2.87544996400288, |
| "grad_norm": 90.92398834228516, |
| "learning_rate": 1.2931034482758623e-06, |
| "loss": 20.5969873046875, |
| "step": 1000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 2.579054832458496, |
| "eval_model_preparation_time": 0.0127, |
| "eval_rouge1": 43.6198, |
| "eval_rouge2": 21.2562, |
| "eval_rougeL": 37.0629, |
| "eval_rougeLsum": 36.9731, |
| "eval_runtime": 178.3925, |
| "eval_samples_per_second": 1.945, |
| "eval_steps_per_second": 1.945, |
| "step": 1044 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 1044, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4686367027249152.0, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
