| { | |
| "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": 5.490695476531982, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5637, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7192982456140351, | |
| "eval_f1": 0.6312141018023371, | |
| "eval_loss": 0.504301905632019, | |
| "eval_precision": 0.6505857648099027, | |
| "eval_recall": 0.6238861611202037, | |
| "eval_runtime": 4.8371, | |
| "eval_samples_per_second": 82.488, | |
| "eval_steps_per_second": 10.337, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.642432689666748, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5058, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7343358395989975, | |
| "eval_f1": 0.7022388059701492, | |
| "eval_loss": 0.4904969036579132, | |
| "eval_precision": 0.695631275368511, | |
| "eval_recall": 0.7220403709765412, | |
| "eval_runtime": 5.048, | |
| "eval_samples_per_second": 79.041, | |
| "eval_steps_per_second": 9.905, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.178711891174316, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4607, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7844611528822055, | |
| "eval_f1": 0.7495328467153285, | |
| "eval_loss": 0.42068806290626526, | |
| "eval_precision": 0.7424812030075187, | |
| "eval_recall": 0.7600018185124569, | |
| "eval_runtime": 5.0555, | |
| "eval_samples_per_second": 78.924, | |
| "eval_steps_per_second": 9.89, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.523857355117798, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3992, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8148090847205891, | |
| "eval_loss": 0.37228837609291077, | |
| "eval_precision": 0.8221132552989561, | |
| "eval_recall": 0.8086015639207129, | |
| "eval_runtime": 5.059, | |
| "eval_samples_per_second": 78.87, | |
| "eval_steps_per_second": 9.883, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.291674852371216, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3565, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8145363408521303, | |
| "eval_f1": 0.7872030673431734, | |
| "eval_loss": 0.3854798376560211, | |
| "eval_precision": 0.7773381294964029, | |
| "eval_recall": 0.8037825059101655, | |
| "eval_runtime": 5.0549, | |
| "eval_samples_per_second": 78.933, | |
| "eval_steps_per_second": 9.891, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 5.28938102722168, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.332, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8270676691729323, | |
| "eval_f1": 0.7977104565126344, | |
| "eval_loss": 0.36892175674438477, | |
| "eval_precision": 0.7902630826422647, | |
| "eval_recall": 0.8076468448808873, | |
| "eval_runtime": 5.0865, | |
| "eval_samples_per_second": 78.443, | |
| "eval_steps_per_second": 9.83, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 3.7372069358825684, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3089, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8446115288220551, | |
| "eval_f1": 0.8116089923236262, | |
| "eval_loss": 0.3518911898136139, | |
| "eval_precision": 0.8132271892222902, | |
| "eval_recall": 0.810056373886161, | |
| "eval_runtime": 5.0912, | |
| "eval_samples_per_second": 78.37, | |
| "eval_steps_per_second": 9.821, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 6.950363636016846, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2979, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8263588263588264, | |
| "eval_loss": 0.3406260311603546, | |
| "eval_precision": 0.8289473684210527, | |
| "eval_recall": 0.8239225313693399, | |
| "eval_runtime": 5.0511, | |
| "eval_samples_per_second": 78.993, | |
| "eval_steps_per_second": 9.899, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 5.673851013183594, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2887, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8203564970291914, | |
| "eval_loss": 0.35820019245147705, | |
| "eval_precision": 0.8132495387453875, | |
| "eval_recall": 0.8293326059283506, | |
| "eval_runtime": 5.0494, | |
| "eval_samples_per_second": 79.019, | |
| "eval_steps_per_second": 9.902, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.3661675453186035, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.268, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8315822595375324, | |
| "eval_loss": 0.33940380811691284, | |
| "eval_precision": 0.8361280487804879, | |
| "eval_recall": 0.82746863066012, | |
| "eval_runtime": 5.053, | |
| "eval_samples_per_second": 78.963, | |
| "eval_steps_per_second": 9.895, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.5574766397476196, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.267, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8272399255573702, | |
| "eval_loss": 0.3338986337184906, | |
| "eval_precision": 0.8280735957109784, | |
| "eval_recall": 0.8264229859974541, | |
| "eval_runtime": 5.0604, | |
| "eval_samples_per_second": 78.848, | |
| "eval_steps_per_second": 9.881, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 17.53325843811035, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2609, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8365204824303285, | |
| "eval_loss": 0.339706152677536, | |
| "eval_precision": 0.8313636363636363, | |
| "eval_recall": 0.8424713584288053, | |
| "eval_runtime": 5.0561, | |
| "eval_samples_per_second": 78.915, | |
| "eval_steps_per_second": 9.889, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.5466238260269165, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2564, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8368501045387564, | |
| "eval_loss": 0.32274091243743896, | |
| "eval_precision": 0.8435514312676942, | |
| "eval_recall": 0.8310147299509002, | |
| "eval_runtime": 5.0671, | |
| "eval_samples_per_second": 78.744, | |
| "eval_steps_per_second": 9.868, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 9.6759033203125, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2566, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8401647707947546, | |
| "eval_loss": 0.3246406316757202, | |
| "eval_precision": 0.8393298751432535, | |
| "eval_recall": 0.8410165484633569, | |
| "eval_runtime": 5.0843, | |
| "eval_samples_per_second": 78.477, | |
| "eval_steps_per_second": 9.834, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.8860002756118774, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2503, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8484099018899409, | |
| "eval_loss": 0.32968148589134216, | |
| "eval_precision": 0.8430645161290322, | |
| "eval_recall": 0.8545644662665939, | |
| "eval_runtime": 5.0647, | |
| "eval_samples_per_second": 78.781, | |
| "eval_steps_per_second": 9.872, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 3.0858819484710693, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2539, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8450716845878137, | |
| "eval_loss": 0.3228394687175751, | |
| "eval_precision": 0.8404471544715447, | |
| "eval_recall": 0.8502909619930896, | |
| "eval_runtime": 5.0463, | |
| "eval_samples_per_second": 79.068, | |
| "eval_steps_per_second": 9.908, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 6.341916084289551, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2478, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8263588263588264, | |
| "eval_loss": 0.31420159339904785, | |
| "eval_precision": 0.8289473684210527, | |
| "eval_recall": 0.8239225313693399, | |
| "eval_runtime": 5.0557, | |
| "eval_samples_per_second": 78.921, | |
| "eval_steps_per_second": 9.89, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 6.797486305236816, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2449, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8476882658063937, | |
| "eval_loss": 0.31900447607040405, | |
| "eval_precision": 0.8437296561519796, | |
| "eval_recall": 0.8520640116384797, | |
| "eval_runtime": 5.0545, | |
| "eval_samples_per_second": 78.94, | |
| "eval_steps_per_second": 9.892, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.446300506591797, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2401, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8341332527115377, | |
| "eval_loss": 0.31386807560920715, | |
| "eval_precision": 0.8333132275770553, | |
| "eval_recall": 0.8349699945444626, | |
| "eval_runtime": 5.1195, | |
| "eval_samples_per_second": 77.937, | |
| "eval_steps_per_second": 9.767, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 5.507618427276611, | |
| "learning_rate": 0.0, | |
| "loss": 0.2392, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8341332527115377, | |
| "eval_loss": 0.3142927885055542, | |
| "eval_precision": 0.8333132275770553, | |
| "eval_recall": 0.8349699945444626, | |
| "eval_runtime": 5.0576, | |
| "eval_samples_per_second": 78.891, | |
| "eval_steps_per_second": 9.886, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7609911792720000.0, | |
| "train_loss": 0.31493380656007863, | |
| "train_runtime": 1922.5285, | |
| "train_samples_per_second": 37.846, | |
| "train_steps_per_second": 1.269 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7609911792720000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |