| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.585148334503174, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5413, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7368421052631579, | |
| "eval_f1": 0.6531440162271805, | |
| "eval_loss": 0.49492353200912476, | |
| "eval_precision": 0.6762575228471654, | |
| "eval_recall": 0.6437988725222767, | |
| "eval_runtime": 1.7908, | |
| "eval_samples_per_second": 222.81, | |
| "eval_steps_per_second": 27.921, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.6699981689453125, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4306, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8245614035087719, | |
| "eval_f1": 0.8018811712631937, | |
| "eval_loss": 0.3954322338104248, | |
| "eval_precision": 0.7902494331065759, | |
| "eval_recall": 0.825877432260411, | |
| "eval_runtime": 1.8012, | |
| "eval_samples_per_second": 221.517, | |
| "eval_steps_per_second": 27.759, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.988449811935425, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3344, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.8099340368327992, | |
| "eval_loss": 0.33969220519065857, | |
| "eval_precision": 0.8370422043948378, | |
| "eval_recall": 0.7928714311693035, | |
| "eval_runtime": 1.8043, | |
| "eval_samples_per_second": 221.138, | |
| "eval_steps_per_second": 27.712, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.6507456302642822, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2925, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8058239259643384, | |
| "eval_loss": 0.32114899158477783, | |
| "eval_precision": 0.8264119601328903, | |
| "eval_recall": 0.7918257865066376, | |
| "eval_runtime": 1.7976, | |
| "eval_samples_per_second": 221.96, | |
| "eval_steps_per_second": 27.814, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.9831173419952393, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2794, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8365204824303285, | |
| "eval_loss": 0.30638387799263, | |
| "eval_precision": 0.8313636363636363, | |
| "eval_recall": 0.8424713584288053, | |
| "eval_runtime": 1.8026, | |
| "eval_samples_per_second": 221.352, | |
| "eval_steps_per_second": 27.738, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.175257444381714, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2464, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8453047161322887, | |
| "eval_loss": 0.2856903374195099, | |
| "eval_precision": 0.835631596867552, | |
| "eval_recall": 0.8585197308601564, | |
| "eval_runtime": 1.8028, | |
| "eval_samples_per_second": 221.325, | |
| "eval_steps_per_second": 27.735, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.4121947288513184, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2332, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8536612749904566, | |
| "eval_loss": 0.28461554646492004, | |
| "eval_precision": 0.8496330709593418, | |
| "eval_recall": 0.8581105655573741, | |
| "eval_runtime": 1.8, | |
| "eval_samples_per_second": 221.663, | |
| "eval_steps_per_second": 27.777, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.52989387512207, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2216, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8262195121951219, | |
| "eval_loss": 0.2906416654586792, | |
| "eval_precision": 0.8360165151709128, | |
| "eval_recall": 0.8181942171303873, | |
| "eval_runtime": 1.8056, | |
| "eval_samples_per_second": 220.983, | |
| "eval_steps_per_second": 27.692, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 8.88005256652832, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2123, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8386324041811847, | |
| "eval_loss": 0.2781093120574951, | |
| "eval_precision": 0.8487869670976828, | |
| "eval_recall": 0.830287324968176, | |
| "eval_runtime": 1.8334, | |
| "eval_samples_per_second": 217.634, | |
| "eval_steps_per_second": 27.272, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 2.3229901790618896, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1911, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8395201930584144, | |
| "eval_loss": 0.2896389365196228, | |
| "eval_precision": 0.8562091503267973, | |
| "eval_recall": 0.8270594653573378, | |
| "eval_runtime": 1.8031, | |
| "eval_samples_per_second": 221.286, | |
| "eval_steps_per_second": 27.73, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.697758138179779, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1878, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8495838108450449, | |
| "eval_loss": 0.28144657611846924, | |
| "eval_precision": 0.8478991596638655, | |
| "eval_recall": 0.8513366066557555, | |
| "eval_runtime": 1.8197, | |
| "eval_samples_per_second": 219.261, | |
| "eval_steps_per_second": 27.476, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.104419231414795, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1797, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8393634395533442, | |
| "eval_loss": 0.28299498558044434, | |
| "eval_precision": 0.8402278542707444, | |
| "eval_recall": 0.8385160938352427, | |
| "eval_runtime": 1.8064, | |
| "eval_samples_per_second": 220.882, | |
| "eval_steps_per_second": 27.679, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.5454129576683044, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1746, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8332268672959993, | |
| "eval_loss": 0.2900232672691345, | |
| "eval_precision": 0.849623687858982, | |
| "eval_recall": 0.8210129114384433, | |
| "eval_runtime": 1.8092, | |
| "eval_samples_per_second": 220.538, | |
| "eval_steps_per_second": 27.636, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 6.253016471862793, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1677, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.844327731092437, | |
| "eval_loss": 0.27983638644218445, | |
| "eval_precision": 0.8411320530352577, | |
| "eval_recall": 0.8477905073649754, | |
| "eval_runtime": 1.8045, | |
| "eval_samples_per_second": 221.113, | |
| "eval_steps_per_second": 27.708, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.9090829491615295, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1585, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8476882658063937, | |
| "eval_loss": 0.28229421377182007, | |
| "eval_precision": 0.8437296561519796, | |
| "eval_recall": 0.8520640116384797, | |
| "eval_runtime": 1.8037, | |
| "eval_samples_per_second": 221.212, | |
| "eval_steps_per_second": 27.721, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.9292184114456177, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1575, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8511422740140892, | |
| "eval_loss": 0.2815591096878052, | |
| "eval_precision": 0.8412921348314606, | |
| "eval_recall": 0.8645662847790507, | |
| "eval_runtime": 1.8158, | |
| "eval_samples_per_second": 219.738, | |
| "eval_steps_per_second": 27.536, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 4.232359409332275, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.146, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8359175094431583, | |
| "eval_loss": 0.30267345905303955, | |
| "eval_precision": 0.8376607470912432, | |
| "eval_recall": 0.8342425895617385, | |
| "eval_runtime": 1.8565, | |
| "eval_samples_per_second": 214.919, | |
| "eval_steps_per_second": 26.932, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.907904624938965, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1368, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8424651921601347, | |
| "eval_loss": 0.29612866044044495, | |
| "eval_precision": 0.8372140762463343, | |
| "eval_recall": 0.8485179123476996, | |
| "eval_runtime": 1.8011, | |
| "eval_samples_per_second": 221.536, | |
| "eval_steps_per_second": 27.761, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.788642168045044, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.133, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8333016825553572, | |
| "eval_loss": 0.30237412452697754, | |
| "eval_precision": 0.8341507249908615, | |
| "eval_recall": 0.8324695399163484, | |
| "eval_runtime": 1.8062, | |
| "eval_samples_per_second": 220.905, | |
| "eval_steps_per_second": 27.682, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.534482479095459, | |
| "learning_rate": 0.0, | |
| "loss": 0.1377, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8359175094431583, | |
| "eval_loss": 0.30191537737846375, | |
| "eval_precision": 0.8376607470912432, | |
| "eval_recall": 0.8342425895617385, | |
| "eval_runtime": 1.8052, | |
| "eval_samples_per_second": 221.033, | |
| "eval_steps_per_second": 27.698, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.2281026918379987, | |
| "train_runtime": 636.1271, | |
| "train_samples_per_second": 114.38, | |
| "train_steps_per_second": 3.836 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 8444128359504000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |