| { |
| "best_metric": 82.213, |
| "best_model_checkpoint": "/uoa/home/s02sd1/Desktop/Project data/pytorch_project/facebook_26.04/checkpoint-5000", |
| "epoch": 95.23809523809524, |
| "global_step": 8000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 5.95, |
| "learning_rate": 2.8214285714285713e-06, |
| "loss": 0.1312, |
| "step": 500 |
| }, |
| { |
| "epoch": 11.9, |
| "learning_rate": 2.642857142857143e-06, |
| "loss": 0.046, |
| "step": 1000 |
| }, |
| { |
| "epoch": 11.9, |
| "eval_gen_len": 189.9754, |
| "eval_loss": 0.0635865107178688, |
| "eval_rouge1": 81.7351, |
| "eval_rouge2": 74.9502, |
| "eval_rougeL": 65.0568, |
| "eval_rougeLsum": 65.0619, |
| "eval_runtime": 20480.7032, |
| "eval_samples_per_second": 0.01, |
| "eval_steps_per_second": 0.001, |
| "step": 1000 |
| }, |
| { |
| "epoch": 17.86, |
| "learning_rate": 2.4642857142857143e-06, |
| "loss": 0.0227, |
| "step": 1500 |
| }, |
| { |
| "epoch": 23.81, |
| "learning_rate": 2.2857142857142856e-06, |
| "loss": 0.0126, |
| "step": 2000 |
| }, |
| { |
| "epoch": 23.81, |
| "eval_gen_len": 208.1773, |
| "eval_loss": 0.061119239777326584, |
| "eval_rouge1": 82.8965, |
| "eval_rouge2": 76.972, |
| "eval_rougeL": 70.043, |
| "eval_rougeLsum": 70.3743, |
| "eval_runtime": 2676.3375, |
| "eval_samples_per_second": 0.076, |
| "eval_steps_per_second": 0.004, |
| "step": 2000 |
| }, |
| { |
| "epoch": 29.76, |
| "learning_rate": 2.1071428571428572e-06, |
| "loss": 0.0072, |
| "step": 2500 |
| }, |
| { |
| "epoch": 35.71, |
| "learning_rate": 1.928571428571429e-06, |
| "loss": 0.0048, |
| "step": 3000 |
| }, |
| { |
| "epoch": 35.71, |
| "eval_gen_len": 206.1478, |
| "eval_loss": 0.07027211785316467, |
| "eval_rouge1": 84.3357, |
| "eval_rouge2": 79.2414, |
| "eval_rougeL": 72.9218, |
| "eval_rougeLsum": 73.2212, |
| "eval_runtime": 2713.1894, |
| "eval_samples_per_second": 0.075, |
| "eval_steps_per_second": 0.004, |
| "step": 3000 |
| }, |
| { |
| "epoch": 41.67, |
| "learning_rate": 1.7500000000000002e-06, |
| "loss": 0.0033, |
| "step": 3500 |
| }, |
| { |
| "epoch": 47.62, |
| "learning_rate": 1.5714285714285714e-06, |
| "loss": 0.0025, |
| "step": 4000 |
| }, |
| { |
| "epoch": 47.62, |
| "eval_gen_len": 210.4483, |
| "eval_loss": 0.06871643662452698, |
| "eval_rouge1": 83.8269, |
| "eval_rouge2": 78.3266, |
| "eval_rougeL": 72.7844, |
| "eval_rougeLsum": 73.1385, |
| "eval_runtime": 3070.1132, |
| "eval_samples_per_second": 0.066, |
| "eval_steps_per_second": 0.004, |
| "step": 4000 |
| }, |
| { |
| "epoch": 53.57, |
| "learning_rate": 1.392857142857143e-06, |
| "loss": 0.0018, |
| "step": 4500 |
| }, |
| { |
| "epoch": 59.52, |
| "learning_rate": 1.2142857142857144e-06, |
| "loss": 0.0016, |
| "step": 5000 |
| }, |
| { |
| "epoch": 59.52, |
| "eval_gen_len": 199.7389, |
| "eval_loss": 0.07927948236465454, |
| "eval_rouge1": 86.8825, |
| "eval_rouge2": 82.213, |
| "eval_rougeL": 77.2112, |
| "eval_rougeLsum": 77.4106, |
| "eval_runtime": 2247.4623, |
| "eval_samples_per_second": 0.09, |
| "eval_steps_per_second": 0.005, |
| "step": 5000 |
| }, |
| { |
| "epoch": 65.48, |
| "learning_rate": 1.0357142857142857e-06, |
| "loss": 0.0012, |
| "step": 5500 |
| }, |
| { |
| "epoch": 71.43, |
| "learning_rate": 8.571428571428571e-07, |
| "loss": 0.001, |
| "step": 6000 |
| }, |
| { |
| "epoch": 71.43, |
| "eval_gen_len": 202.665, |
| "eval_loss": 0.08206681907176971, |
| "eval_rouge1": 86.0452, |
| "eval_rouge2": 81.3296, |
| "eval_rougeL": 75.9829, |
| "eval_rougeLsum": 76.3571, |
| "eval_runtime": 2123.8856, |
| "eval_samples_per_second": 0.096, |
| "eval_steps_per_second": 0.005, |
| "step": 6000 |
| }, |
| { |
| "epoch": 77.38, |
| "learning_rate": 6.785714285714286e-07, |
| "loss": 0.0009, |
| "step": 6500 |
| }, |
| { |
| "epoch": 83.33, |
| "learning_rate": 5e-07, |
| "loss": 0.0008, |
| "step": 7000 |
| }, |
| { |
| "epoch": 83.33, |
| "eval_gen_len": 205.0788, |
| "eval_loss": 0.08218137919902802, |
| "eval_rouge1": 85.9623, |
| "eval_rouge2": 81.1543, |
| "eval_rougeL": 76.8811, |
| "eval_rougeLsum": 77.0526, |
| "eval_runtime": 2090.4257, |
| "eval_samples_per_second": 0.097, |
| "eval_steps_per_second": 0.005, |
| "step": 7000 |
| }, |
| { |
| "epoch": 89.29, |
| "learning_rate": 3.214285714285714e-07, |
| "loss": 0.0007, |
| "step": 7500 |
| }, |
| { |
| "epoch": 95.24, |
| "learning_rate": 1.4285714285714285e-07, |
| "loss": 0.0006, |
| "step": 8000 |
| }, |
| { |
| "epoch": 95.24, |
| "eval_gen_len": 206.7488, |
| "eval_loss": 0.07930376380681992, |
| "eval_rouge1": 85.5313, |
| "eval_rouge2": 80.6524, |
| "eval_rougeL": 75.6534, |
| "eval_rougeLsum": 75.8372, |
| "eval_runtime": 2147.6107, |
| "eval_samples_per_second": 0.095, |
| "eval_steps_per_second": 0.005, |
| "step": 8000 |
| } |
| ], |
| "max_steps": 8400, |
| "num_train_epochs": 100, |
| "total_flos": 1.1606068796537242e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
