| { | |
| "best_metric": 0.51416015625, | |
| "best_model_checkpoint": "autotrain-jvq6k-yf3ca/checkpoint-570", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 570, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 8.074106342959622, | |
| "learning_rate": 4.210526315789474e-05, | |
| "loss": 4.7837, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 2.5780648325169624, | |
| "learning_rate": 0.0001394736842105263, | |
| "loss": 1.6204, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 2.533813969716365, | |
| "learning_rate": 0.00023684210526315788, | |
| "loss": 0.8889, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 2.3405358212713185, | |
| "learning_rate": 0.00033421052631578944, | |
| "loss": 0.8323, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 1.6021401005483846, | |
| "learning_rate": 0.0004315789473684211, | |
| "loss": 0.7663, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 9.6497, | |
| "eval_loss": 0.5634765625, | |
| "eval_rouge1": 84.8501, | |
| "eval_rouge2": 72.9759, | |
| "eval_rougeL": 83.9381, | |
| "eval_rougeLsum": 83.9882, | |
| "eval_runtime": 59.9525, | |
| "eval_samples_per_second": 25.27, | |
| "eval_steps_per_second": 0.4, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 2.014072063970188, | |
| "learning_rate": 0.0004967836257309941, | |
| "loss": 0.6051, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 2.5322811119510225, | |
| "learning_rate": 0.0004868421052631579, | |
| "loss": 0.6124, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 1.684844161543083, | |
| "learning_rate": 0.00047690058479532164, | |
| "loss": 0.5759, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 1.830769701886568, | |
| "learning_rate": 0.00046637426900584796, | |
| "loss": 0.6172, | |
| "step": 333 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 1.482259801017781, | |
| "learning_rate": 0.00045584795321637427, | |
| "loss": 0.5658, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 9.709, | |
| "eval_loss": 0.52587890625, | |
| "eval_rouge1": 86.3194, | |
| "eval_rouge2": 74.6858, | |
| "eval_rougeL": 85.4633, | |
| "eval_rougeLsum": 85.4901, | |
| "eval_runtime": 58.3784, | |
| "eval_samples_per_second": 25.951, | |
| "eval_steps_per_second": 0.411, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 1.4815030414040293, | |
| "learning_rate": 0.0004450292397660819, | |
| "loss": 0.4243, | |
| "step": 407 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 1.4036378645480725, | |
| "learning_rate": 0.0004342105263157895, | |
| "loss": 0.3784, | |
| "step": 444 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 1.5836309025847368, | |
| "learning_rate": 0.0004233918128654971, | |
| "loss": 0.3789, | |
| "step": 481 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 1.6685671897157268, | |
| "learning_rate": 0.0004125730994152047, | |
| "loss": 0.4048, | |
| "step": 518 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "grad_norm": 1.9497904549926495, | |
| "learning_rate": 0.0004017543859649123, | |
| "loss": 0.3807, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 9.8561, | |
| "eval_loss": 0.51416015625, | |
| "eval_rouge1": 87.4319, | |
| "eval_rouge2": 76.4229, | |
| "eval_rougeL": 86.4987, | |
| "eval_rougeLsum": 86.5222, | |
| "eval_runtime": 59.8804, | |
| "eval_samples_per_second": 25.3, | |
| "eval_steps_per_second": 0.401, | |
| "step": 570 | |
| } | |
| ], | |
| "logging_steps": 37, | |
| "max_steps": 1900, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 438831341568.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |