| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 10560, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.3234807252883911, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.7663, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8658948623670873, | |
| "eval_f1": 0.17383431416930736, | |
| "eval_loss": 0.43804919719696045, | |
| "eval_precision": 0.39344262295081966, | |
| "eval_recall": 0.1115630447414294, | |
| "eval_runtime": 4.4497, | |
| "eval_samples_per_second": 210.127, | |
| "eval_steps_per_second": 3.371, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.9100218415260315, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.3481, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9338666401669482, | |
| "eval_f1": 0.6638874413757164, | |
| "eval_loss": 0.22202354669570923, | |
| "eval_precision": 0.6017949929145017, | |
| "eval_recall": 0.7402672864613596, | |
| "eval_runtime": 4.5326, | |
| "eval_samples_per_second": 206.282, | |
| "eval_steps_per_second": 3.309, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.4061142206192017, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.2139, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9400278247043625, | |
| "eval_f1": 0.7339308578745198, | |
| "eval_loss": 0.17896419763565063, | |
| "eval_precision": 0.6561355311355311, | |
| "eval_recall": 0.8326554328878559, | |
| "eval_runtime": 4.5296, | |
| "eval_samples_per_second": 206.418, | |
| "eval_steps_per_second": 3.312, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.0871784687042236, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1777, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9511577064493689, | |
| "eval_f1": 0.7799841143764892, | |
| "eval_loss": 0.1534842550754547, | |
| "eval_precision": 0.7164396887159533, | |
| "eval_recall": 0.8558977338756537, | |
| "eval_runtime": 4.531, | |
| "eval_samples_per_second": 206.356, | |
| "eval_steps_per_second": 3.311, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.1403982639312744, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.1578, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9535426811090132, | |
| "eval_f1": 0.797761790567546, | |
| "eval_loss": 0.1444850116968155, | |
| "eval_precision": 0.7367125984251969, | |
| "eval_recall": 0.8698431144683324, | |
| "eval_runtime": 4.519, | |
| "eval_samples_per_second": 206.903, | |
| "eval_steps_per_second": 3.319, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.446531057357788, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.1469, | |
| "step": 3168 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9535426811090132, | |
| "eval_f1": 0.7861060329067641, | |
| "eval_loss": 0.1441143900156021, | |
| "eval_precision": 0.7139468690702088, | |
| "eval_recall": 0.8744915746658919, | |
| "eval_runtime": 4.4888, | |
| "eval_samples_per_second": 208.297, | |
| "eval_steps_per_second": 3.342, | |
| "step": 3168 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.257020950317383, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.1399, | |
| "step": 3696 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9523501937791911, | |
| "eval_f1": 0.7919812548815414, | |
| "eval_loss": 0.14534059166908264, | |
| "eval_precision": 0.7174528301886792, | |
| "eval_recall": 0.8837884950610111, | |
| "eval_runtime": 4.5376, | |
| "eval_samples_per_second": 206.054, | |
| "eval_steps_per_second": 3.306, | |
| "step": 3696 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.8136665225028992, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1333, | |
| "step": 4224 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9547351684388353, | |
| "eval_f1": 0.7994743758212879, | |
| "eval_loss": 0.14028875529766083, | |
| "eval_precision": 0.7298464491362764, | |
| "eval_recall": 0.8837884950610111, | |
| "eval_runtime": 4.5281, | |
| "eval_samples_per_second": 206.486, | |
| "eval_steps_per_second": 3.313, | |
| "step": 4224 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.5226961374282837, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.1273, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9563251515452649, | |
| "eval_f1": 0.7998934185984545, | |
| "eval_loss": 0.13680371642112732, | |
| "eval_precision": 0.7386811023622047, | |
| "eval_recall": 0.8721673445671121, | |
| "eval_runtime": 4.8276, | |
| "eval_samples_per_second": 193.679, | |
| "eval_steps_per_second": 3.107, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.3885184526443481, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1246, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9569213952101759, | |
| "eval_f1": 0.8041566746602716, | |
| "eval_loss": 0.1341899335384369, | |
| "eval_precision": 0.7426181102362205, | |
| "eval_recall": 0.8768158047646717, | |
| "eval_runtime": 4.8281, | |
| "eval_samples_per_second": 193.657, | |
| "eval_steps_per_second": 3.107, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.2650506496429443, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1195, | |
| "step": 5808 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9571201430984796, | |
| "eval_f1": 0.8011649457241197, | |
| "eval_loss": 0.13513582944869995, | |
| "eval_precision": 0.7358949416342413, | |
| "eval_recall": 0.8791400348634515, | |
| "eval_runtime": 4.5526, | |
| "eval_samples_per_second": 205.379, | |
| "eval_steps_per_second": 3.295, | |
| "step": 5808 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.5340452194213867, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1172, | |
| "step": 6336 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9573188909867832, | |
| "eval_f1": 0.8020143122183939, | |
| "eval_loss": 0.13494175672531128, | |
| "eval_precision": 0.7373294346978557, | |
| "eval_recall": 0.8791400348634515, | |
| "eval_runtime": 4.563, | |
| "eval_samples_per_second": 204.908, | |
| "eval_steps_per_second": 3.287, | |
| "step": 6336 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.7720882892608643, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1155, | |
| "step": 6864 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.958113882539998, | |
| "eval_f1": 0.805014670578821, | |
| "eval_loss": 0.12961605191230774, | |
| "eval_precision": 0.7440828402366864, | |
| "eval_recall": 0.8768158047646717, | |
| "eval_runtime": 4.5551, | |
| "eval_samples_per_second": 205.265, | |
| "eval_steps_per_second": 3.293, | |
| "step": 6864 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 1.088473916053772, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1118, | |
| "step": 7392 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9577163867633907, | |
| "eval_f1": 0.797761790567546, | |
| "eval_loss": 0.13024814426898956, | |
| "eval_precision": 0.7367125984251969, | |
| "eval_recall": 0.8698431144683324, | |
| "eval_runtime": 4.5273, | |
| "eval_samples_per_second": 206.527, | |
| "eval_steps_per_second": 3.313, | |
| "step": 7392 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.3462263345718384, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1111, | |
| "step": 7920 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9577163867633907, | |
| "eval_f1": 0.8041566746602716, | |
| "eval_loss": 0.13221099972724915, | |
| "eval_precision": 0.7426181102362205, | |
| "eval_recall": 0.8768158047646717, | |
| "eval_runtime": 4.5319, | |
| "eval_samples_per_second": 206.317, | |
| "eval_steps_per_second": 3.31, | |
| "step": 7920 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.848177433013916, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1097, | |
| "step": 8448 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9577163867633907, | |
| "eval_f1": 0.7969124301304232, | |
| "eval_loss": 0.1302962750196457, | |
| "eval_precision": 0.7352652259332023, | |
| "eval_recall": 0.8698431144683324, | |
| "eval_runtime": 4.6208, | |
| "eval_samples_per_second": 202.344, | |
| "eval_steps_per_second": 3.246, | |
| "step": 8448 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 1.2858753204345703, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1094, | |
| "step": 8976 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9573188909867832, | |
| "eval_f1": 0.7973439575033201, | |
| "eval_loss": 0.13064691424369812, | |
| "eval_precision": 0.7343444227005871, | |
| "eval_recall": 0.8721673445671121, | |
| "eval_runtime": 4.5402, | |
| "eval_samples_per_second": 205.938, | |
| "eval_steps_per_second": 3.304, | |
| "step": 8976 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 1.495869755744934, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1077, | |
| "step": 9504 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.9577163867633907, | |
| "eval_f1": 0.7990417886611658, | |
| "eval_loss": 0.13186967372894287, | |
| "eval_precision": 0.7372298624754421, | |
| "eval_recall": 0.8721673445671121, | |
| "eval_runtime": 4.5414, | |
| "eval_samples_per_second": 205.882, | |
| "eval_steps_per_second": 3.303, | |
| "step": 9504 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.5424643754959106, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1065, | |
| "step": 10032 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9577163867633907, | |
| "eval_f1": 0.7973297730307075, | |
| "eval_loss": 0.12963254749774933, | |
| "eval_precision": 0.7376482213438735, | |
| "eval_recall": 0.8675188843695526, | |
| "eval_runtime": 4.5053, | |
| "eval_samples_per_second": 207.533, | |
| "eval_steps_per_second": 3.329, | |
| "step": 10032 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.9210747480392456, | |
| "learning_rate": 0.0, | |
| "loss": 0.1078, | |
| "step": 10560 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9577163867633907, | |
| "eval_f1": 0.7951935914552738, | |
| "eval_loss": 0.13010983169078827, | |
| "eval_precision": 0.7356719367588933, | |
| "eval_recall": 0.8651946542707728, | |
| "eval_runtime": 4.5258, | |
| "eval_samples_per_second": 206.594, | |
| "eval_steps_per_second": 3.314, | |
| "step": 10560 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 10560, | |
| "total_flos": 4545096690358590.0, | |
| "train_loss": 0.17260197039806482, | |
| "train_runtime": 1230.9591, | |
| "train_samples_per_second": 137.08, | |
| "train_steps_per_second": 8.579 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10560, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 4545096690358590.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |