| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.3741588156123825, | |
| "eval_steps": 500, | |
| "global_step": 6500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.8654104979811575e-05, | |
| "loss": 0.2266, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.730820995962315e-05, | |
| "loss": 0.1566, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.148934468626976, | |
| "eval_runtime": 122.3258, | |
| "eval_samples_per_second": 12.156, | |
| "eval_steps_per_second": 3.041, | |
| "step": 1486 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 1.5962314939434724e-05, | |
| "loss": 0.136, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 1.46164199192463e-05, | |
| "loss": 0.1165, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 1.3270524899057875e-05, | |
| "loss": 0.1189, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.11046190559864044, | |
| "eval_runtime": 122.3143, | |
| "eval_samples_per_second": 12.157, | |
| "eval_steps_per_second": 3.041, | |
| "step": 2972 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1.192462987886945e-05, | |
| "loss": 0.1072, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 1.0578734858681023e-05, | |
| "loss": 0.1073, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 9.2328398384926e-06, | |
| "loss": 0.0775, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.13019011914730072, | |
| "eval_runtime": 122.3055, | |
| "eval_samples_per_second": 12.158, | |
| "eval_steps_per_second": 3.042, | |
| "step": 4458 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 7.886944818304172e-06, | |
| "loss": 0.0891, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 6.541049798115747e-06, | |
| "loss": 0.0755, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 5.195154777927323e-06, | |
| "loss": 0.0802, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.12089045345783234, | |
| "eval_runtime": 122.2813, | |
| "eval_samples_per_second": 12.16, | |
| "eval_steps_per_second": 3.042, | |
| "step": 5944 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 3.849259757738897e-06, | |
| "loss": 0.0734, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 2.503364737550471e-06, | |
| "loss": 0.0759, | |
| "step": 6500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7430, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 1.17038997504e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |