| { | |
| "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": 5.091732025146484, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5658, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7268170426065163, | |
| "eval_f1": 0.6550441396924102, | |
| "eval_loss": 0.519511878490448, | |
| "eval_precision": 0.6646488294314381, | |
| "eval_recall": 0.6492089470812875, | |
| "eval_runtime": 5.1503, | |
| "eval_samples_per_second": 77.471, | |
| "eval_steps_per_second": 9.708, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.045337200164795, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5125, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7293233082706767, | |
| "eval_f1": 0.6854598540145985, | |
| "eval_loss": 0.5060133337974548, | |
| "eval_precision": 0.6804511278195489, | |
| "eval_recall": 0.6934897254046191, | |
| "eval_runtime": 5.0488, | |
| "eval_samples_per_second": 79.029, | |
| "eval_steps_per_second": 9.903, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.02074670791626, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4809, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7669172932330827, | |
| "eval_f1": 0.7166907166907166, | |
| "eval_loss": 0.46860969066619873, | |
| "eval_precision": 0.718421052631579, | |
| "eval_recall": 0.7150845608292418, | |
| "eval_runtime": 5.0676, | |
| "eval_samples_per_second": 78.735, | |
| "eval_steps_per_second": 9.867, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.456829786300659, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4353, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7919799498746867, | |
| "eval_f1": 0.7417274322480016, | |
| "eval_loss": 0.4295312464237213, | |
| "eval_precision": 0.7499839010882864, | |
| "eval_recall": 0.7353155119112567, | |
| "eval_runtime": 5.071, | |
| "eval_samples_per_second": 78.683, | |
| "eval_steps_per_second": 9.86, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.0160815715789795, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4116, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8020050125313283, | |
| "eval_f1": 0.7713964535663778, | |
| "eval_loss": 0.4171212613582611, | |
| "eval_precision": 0.7628187206441512, | |
| "eval_recall": 0.7849154391707583, | |
| "eval_runtime": 5.0771, | |
| "eval_samples_per_second": 78.588, | |
| "eval_steps_per_second": 9.848, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.68072509765625, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3809, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8446115288220551, | |
| "eval_f1": 0.8096491228070175, | |
| "eval_loss": 0.38647741079330444, | |
| "eval_precision": 0.8148460960960962, | |
| "eval_recall": 0.8050554646299327, | |
| "eval_runtime": 5.0568, | |
| "eval_samples_per_second": 78.904, | |
| "eval_steps_per_second": 9.888, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.0169405937194824, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3681, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8176861216035092, | |
| "eval_loss": 0.36971357464790344, | |
| "eval_precision": 0.8193355786895284, | |
| "eval_recall": 0.8161029278050556, | |
| "eval_runtime": 5.0528, | |
| "eval_samples_per_second": 78.966, | |
| "eval_steps_per_second": 9.895, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.219416618347168, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3469, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8101852212906999, | |
| "eval_loss": 0.35537657141685486, | |
| "eval_precision": 0.820642024599137, | |
| "eval_recall": 0.8018276050190944, | |
| "eval_runtime": 5.0531, | |
| "eval_samples_per_second": 78.961, | |
| "eval_steps_per_second": 9.895, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 5.871381759643555, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3455, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8157894736842105, | |
| "eval_loss": 0.3493554890155792, | |
| "eval_precision": 0.8210867117117118, | |
| "eval_recall": 0.8111020185488271, | |
| "eval_runtime": 5.0737, | |
| "eval_samples_per_second": 78.641, | |
| "eval_steps_per_second": 9.855, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 6.4938273429870605, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3284, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8095647193585338, | |
| "eval_loss": 0.34365448355674744, | |
| "eval_precision": 0.8289393939393939, | |
| "eval_recall": 0.7960992907801419, | |
| "eval_runtime": 5.0561, | |
| "eval_samples_per_second": 78.915, | |
| "eval_steps_per_second": 9.889, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 4.7792277336120605, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.3132, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8242843661528783, | |
| "eval_loss": 0.3370875418186188, | |
| "eval_precision": 0.8389366308055628, | |
| "eval_recall": 0.8131933078741589, | |
| "eval_runtime": 5.0782, | |
| "eval_samples_per_second": 78.572, | |
| "eval_steps_per_second": 9.846, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 11.339967727661133, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3042, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8237632508833923, | |
| "eval_loss": 0.3370998501777649, | |
| "eval_precision": 0.8254439681567667, | |
| "eval_recall": 0.8221494817239499, | |
| "eval_runtime": 5.0538, | |
| "eval_samples_per_second": 78.95, | |
| "eval_steps_per_second": 9.893, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 3.7291290760040283, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.3063, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8232837709585574, | |
| "eval_loss": 0.331680566072464, | |
| "eval_precision": 0.8405874144461426, | |
| "eval_recall": 0.8106928532460447, | |
| "eval_runtime": 5.0832, | |
| "eval_samples_per_second": 78.494, | |
| "eval_steps_per_second": 9.836, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 9.840410232543945, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.3013, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8306935047100303, | |
| "eval_loss": 0.33042553067207336, | |
| "eval_precision": 0.8372758729160114, | |
| "eval_recall": 0.8249681760320058, | |
| "eval_runtime": 5.0615, | |
| "eval_samples_per_second": 78.83, | |
| "eval_steps_per_second": 9.878, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.2072865962982178, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2928, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8289650949173301, | |
| "eval_loss": 0.3295370638370514, | |
| "eval_precision": 0.8325081997648369, | |
| "eval_recall": 0.82569558101473, | |
| "eval_runtime": 5.0623, | |
| "eval_samples_per_second": 78.818, | |
| "eval_steps_per_second": 9.877, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 7.374602317810059, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2864, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8324514991181657, | |
| "eval_loss": 0.32842549681663513, | |
| "eval_precision": 0.8350877192982455, | |
| "eval_recall": 0.8299690852882342, | |
| "eval_runtime": 5.0572, | |
| "eval_samples_per_second": 78.897, | |
| "eval_steps_per_second": 9.887, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 5.933995723724365, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2819, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8271551457392166, | |
| "eval_loss": 0.3254449963569641, | |
| "eval_precision": 0.8347358430876305, | |
| "eval_recall": 0.8206946717585015, | |
| "eval_runtime": 5.0559, | |
| "eval_samples_per_second": 78.917, | |
| "eval_steps_per_second": 9.889, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 4.023169994354248, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2877, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8280701754385965, | |
| "eval_loss": 0.32487839460372925, | |
| "eval_precision": 0.833567942942943, | |
| "eval_recall": 0.8231951263866157, | |
| "eval_runtime": 5.0643, | |
| "eval_samples_per_second": 78.787, | |
| "eval_steps_per_second": 9.873, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.3332839012146, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2819, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8333281762485303, | |
| "eval_loss": 0.3241012394428253, | |
| "eval_precision": 0.8410471369819678, | |
| "eval_recall": 0.8267412256773959, | |
| "eval_runtime": 5.0908, | |
| "eval_samples_per_second": 78.377, | |
| "eval_steps_per_second": 9.822, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 7.439182758331299, | |
| "learning_rate": 0.0, | |
| "loss": 0.2803, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8306935047100303, | |
| "eval_loss": 0.3238792419433594, | |
| "eval_precision": 0.8372758729160114, | |
| "eval_recall": 0.8249681760320058, | |
| "eval_runtime": 5.0792, | |
| "eval_samples_per_second": 78.556, | |
| "eval_steps_per_second": 9.844, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7597037114448000.0, | |
| "train_loss": 0.35560139046340694, | |
| "train_runtime": 1953.7377, | |
| "train_samples_per_second": 37.241, | |
| "train_steps_per_second": 1.249 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7597037114448000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |