| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "global_step": 5530, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 9.64737793851718e-06, | |
| "loss": 0.4565, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 9.285714285714288e-06, | |
| "loss": 0.2794, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.8460078451221056, | |
| "eval_loss": 0.23727749288082123, | |
| "eval_runtime": 1.2571, | |
| "eval_samples_per_second": 3128.599, | |
| "eval_steps_per_second": 49.319, | |
| "step": 553 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 8.924050632911393e-06, | |
| "loss": 0.2371, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 8.562386980108501e-06, | |
| "loss": 0.2237, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 8.200723327305607e-06, | |
| "loss": 0.208, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.8584722760759984, | |
| "eval_loss": 0.2176441252231598, | |
| "eval_runtime": 1.2612, | |
| "eval_samples_per_second": 3118.527, | |
| "eval_steps_per_second": 49.161, | |
| "step": 1106 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 7.839059674502714e-06, | |
| "loss": 0.2034, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 7.47739602169982e-06, | |
| "loss": 0.1856, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 7.115732368896926e-06, | |
| "loss": 0.1915, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.8542068079640335, | |
| "eval_loss": 0.20573046803474426, | |
| "eval_runtime": 1.2605, | |
| "eval_samples_per_second": 3120.201, | |
| "eval_steps_per_second": 49.187, | |
| "step": 1659 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 6.754068716094033e-06, | |
| "loss": 0.1757, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 6.392405063291139e-06, | |
| "loss": 0.163, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 6.0307414104882465e-06, | |
| "loss": 0.1662, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_f1": 0.8634553628773282, | |
| "eval_loss": 0.22162207961082458, | |
| "eval_runtime": 1.2593, | |
| "eval_samples_per_second": 3123.225, | |
| "eval_steps_per_second": 49.235, | |
| "step": 2212 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 5.669077757685353e-06, | |
| "loss": 0.1514, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 5.30741410488246e-06, | |
| "loss": 0.1472, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_f1": 0.870850622406639, | |
| "eval_loss": 0.21599918603897095, | |
| "eval_runtime": 1.2572, | |
| "eval_samples_per_second": 3128.475, | |
| "eval_steps_per_second": 49.317, | |
| "step": 2765 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 4.9457504520795664e-06, | |
| "loss": 0.1502, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "learning_rate": 4.584086799276673e-06, | |
| "loss": 0.1321, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 4.22242314647378e-06, | |
| "loss": 0.132, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_f1": 0.8702744691869497, | |
| "eval_loss": 0.2296592891216278, | |
| "eval_runtime": 1.2604, | |
| "eval_samples_per_second": 3120.552, | |
| "eval_steps_per_second": 49.193, | |
| "step": 3318 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 3.860759493670886e-06, | |
| "loss": 0.1311, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 3.499095840867993e-06, | |
| "loss": 0.1204, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 3.1374321880650997e-06, | |
| "loss": 0.1255, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_f1": 0.8709216170544651, | |
| "eval_loss": 0.2616898715496063, | |
| "eval_runtime": 1.259, | |
| "eval_samples_per_second": 3123.885, | |
| "eval_steps_per_second": 49.245, | |
| "step": 3871 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 2.7757685352622067e-06, | |
| "loss": 0.1178, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 2.414104882459313e-06, | |
| "loss": 0.1067, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 2.0524412296564196e-06, | |
| "loss": 0.1162, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_f1": 0.8737610850286907, | |
| "eval_loss": 0.2972545921802521, | |
| "eval_runtime": 1.2617, | |
| "eval_samples_per_second": 3117.195, | |
| "eval_steps_per_second": 49.14, | |
| "step": 4424 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 1.6907775768535265e-06, | |
| "loss": 0.0998, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 8.68, | |
| "learning_rate": 1.3291139240506329e-06, | |
| "loss": 0.1036, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_f1": 0.8713283077722902, | |
| "eval_loss": 0.2818024754524231, | |
| "eval_runtime": 1.2619, | |
| "eval_samples_per_second": 3116.737, | |
| "eval_steps_per_second": 49.132, | |
| "step": 4977 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 9.692585895117542e-07, | |
| "loss": 0.1046, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "learning_rate": 6.075949367088608e-07, | |
| "loss": 0.0923, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 2.4593128390596746e-07, | |
| "loss": 0.1, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_f1": 0.8746249184605349, | |
| "eval_loss": 0.3006608486175537, | |
| "eval_runtime": 1.2724, | |
| "eval_samples_per_second": 3090.907, | |
| "eval_steps_per_second": 48.725, | |
| "step": 5530 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 5530, | |
| "total_flos": 2.160884009653253e+16, | |
| "train_loss": 0.16197810992194342, | |
| "train_runtime": 673.3062, | |
| "train_samples_per_second": 525.63, | |
| "train_steps_per_second": 8.213 | |
| } | |
| ], | |
| "max_steps": 5530, | |
| "num_train_epochs": 10, | |
| "total_flos": 2.160884009653253e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |