| { | |
| "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": 4.3284077644348145, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5535, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7268170426065163, | |
| "eval_f1": 0.6326680574676724, | |
| "eval_loss": 0.5077849626541138, | |
| "eval_precision": 0.6605799373040753, | |
| "eval_recall": 0.6242044008001455, | |
| "eval_runtime": 5.0717, | |
| "eval_samples_per_second": 78.673, | |
| "eval_steps_per_second": 9.859, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.732577085494995, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4682, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8170426065162907, | |
| "eval_f1": 0.777617444284111, | |
| "eval_loss": 0.4184626042842865, | |
| "eval_precision": 0.7798245614035089, | |
| "eval_recall": 0.7755501000181851, | |
| "eval_runtime": 5.0889, | |
| "eval_samples_per_second": 78.406, | |
| "eval_steps_per_second": 9.825, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.4716732501983643, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3849, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8170426065162907, | |
| "eval_f1": 0.757268931723293, | |
| "eval_loss": 0.38087406754493713, | |
| "eval_precision": 0.7968253968253969, | |
| "eval_recall": 0.7380432805964721, | |
| "eval_runtime": 5.1872, | |
| "eval_samples_per_second": 76.92, | |
| "eval_steps_per_second": 9.639, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.9364979267120361, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3127, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8289446964056049, | |
| "eval_loss": 0.32795679569244385, | |
| "eval_precision": 0.8266129032258065, | |
| "eval_recall": 0.8314238952536825, | |
| "eval_runtime": 5.0897, | |
| "eval_samples_per_second": 78.394, | |
| "eval_steps_per_second": 9.824, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.656118392944336, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2869, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8341332527115377, | |
| "eval_loss": 0.3168599307537079, | |
| "eval_precision": 0.8333132275770553, | |
| "eval_recall": 0.8349699945444626, | |
| "eval_runtime": 5.064, | |
| "eval_samples_per_second": 78.791, | |
| "eval_steps_per_second": 9.874, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.3921332359313965, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.274, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8576006759069754, | |
| "eval_loss": 0.3217551112174988, | |
| "eval_precision": 0.8466769923965081, | |
| "eval_recall": 0.8731132933260592, | |
| "eval_runtime": 5.0567, | |
| "eval_samples_per_second": 78.906, | |
| "eval_steps_per_second": 9.888, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.9414736032485962, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2539, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8417152566223307, | |
| "eval_loss": 0.30381932854652405, | |
| "eval_precision": 0.8378262413446174, | |
| "eval_recall": 0.8460174577195854, | |
| "eval_runtime": 5.0791, | |
| "eval_samples_per_second": 78.557, | |
| "eval_steps_per_second": 9.844, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.456129789352417, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2286, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8341632880321839, | |
| "eval_loss": 0.32023322582244873, | |
| "eval_precision": 0.8479139504563233, | |
| "eval_recall": 0.8235133660665576, | |
| "eval_runtime": 5.2736, | |
| "eval_samples_per_second": 75.66, | |
| "eval_steps_per_second": 9.481, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.962612152099609, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2249, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8662440310793597, | |
| "eval_loss": 0.2973268926143646, | |
| "eval_precision": 0.8606158357771261, | |
| "eval_recall": 0.872704128023277, | |
| "eval_runtime": 5.05, | |
| "eval_samples_per_second": 79.009, | |
| "eval_steps_per_second": 9.901, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.1556286811828613, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2083, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8376560692488731, | |
| "eval_loss": 0.3127811551094055, | |
| "eval_precision": 0.8602278120550546, | |
| "eval_recall": 0.8220585561011093, | |
| "eval_runtime": 5.0596, | |
| "eval_samples_per_second": 78.86, | |
| "eval_steps_per_second": 9.882, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.8982645273208618, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1935, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8721887408091659, | |
| "eval_loss": 0.2957092523574829, | |
| "eval_precision": 0.866466275659824, | |
| "eval_recall": 0.8787506819421713, | |
| "eval_runtime": 5.0454, | |
| "eval_samples_per_second": 79.082, | |
| "eval_steps_per_second": 9.91, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 6.364419460296631, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1859, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8602993213495533, | |
| "eval_loss": 0.2869341969490051, | |
| "eval_precision": 0.8547653958944281, | |
| "eval_recall": 0.8666575741043827, | |
| "eval_runtime": 5.0666, | |
| "eval_samples_per_second": 78.751, | |
| "eval_steps_per_second": 9.869, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.9051835536956787, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1735, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8796992481203008, | |
| "eval_f1": 0.8502252252252251, | |
| "eval_loss": 0.30611610412597656, | |
| "eval_precision": 0.863265306122449, | |
| "eval_recall": 0.8398799781778505, | |
| "eval_runtime": 5.1317, | |
| "eval_samples_per_second": 77.752, | |
| "eval_steps_per_second": 9.743, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 6.753292083740234, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1804, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8897243107769424, | |
| "eval_f1": 0.8695225637671682, | |
| "eval_loss": 0.29550090432167053, | |
| "eval_precision": 0.8631532846715328, | |
| "eval_recall": 0.8769776322967813, | |
| "eval_runtime": 5.0486, | |
| "eval_samples_per_second": 79.032, | |
| "eval_steps_per_second": 9.904, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.1921367347240448, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1628, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8757339815412664, | |
| "eval_loss": 0.2972831130027771, | |
| "eval_precision": 0.8766906299500427, | |
| "eval_recall": 0.8747954173486088, | |
| "eval_runtime": 5.0606, | |
| "eval_samples_per_second": 78.844, | |
| "eval_steps_per_second": 9.88, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.38073158264160156, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1619, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8897243107769424, | |
| "eval_f1": 0.8707140332272888, | |
| "eval_loss": 0.3023494482040405, | |
| "eval_precision": 0.8618432385874246, | |
| "eval_recall": 0.8819785415530097, | |
| "eval_runtime": 5.0599, | |
| "eval_samples_per_second": 78.856, | |
| "eval_steps_per_second": 9.882, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 1.6226332187652588, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1514, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8775533117267087, | |
| "eval_loss": 0.2997310757637024, | |
| "eval_precision": 0.873246730188791, | |
| "eval_recall": 0.8822967812329514, | |
| "eval_runtime": 5.0625, | |
| "eval_samples_per_second": 78.815, | |
| "eval_steps_per_second": 9.877, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.088443756103516, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1503, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8736504011098378, | |
| "eval_loss": 0.3002457618713379, | |
| "eval_precision": 0.8718487394957983, | |
| "eval_recall": 0.8755228223313329, | |
| "eval_runtime": 5.0624, | |
| "eval_samples_per_second": 78.817, | |
| "eval_steps_per_second": 9.877, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 6.321498394012451, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.154, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8730223677032187, | |
| "eval_loss": 0.303114652633667, | |
| "eval_precision": 0.8730223677032187, | |
| "eval_recall": 0.8730223677032187, | |
| "eval_runtime": 5.2549, | |
| "eval_samples_per_second": 75.929, | |
| "eval_steps_per_second": 9.515, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.89798903465271, | |
| "learning_rate": 0.0, | |
| "loss": 0.1408, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8763538792940554, | |
| "eval_loss": 0.30106595158576965, | |
| "eval_precision": 0.8754297605404427, | |
| "eval_recall": 0.877295871976723, | |
| "eval_runtime": 5.0597, | |
| "eval_samples_per_second": 78.858, | |
| "eval_steps_per_second": 9.882, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.24252970335913487, | |
| "train_runtime": 1919.2772, | |
| "train_samples_per_second": 37.91, | |
| "train_steps_per_second": 1.271 | |
| } | |
| ], | |
| "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 | |
| } | |