| { | |
| "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": 5.524814605712891, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.559, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7268170426065163, | |
| "eval_f1": 0.6626643397324026, | |
| "eval_loss": 0.5023075342178345, | |
| "eval_precision": 0.6671439480717831, | |
| "eval_recall": 0.6592107655937443, | |
| "eval_runtime": 5.1011, | |
| "eval_samples_per_second": 78.218, | |
| "eval_steps_per_second": 9.802, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.053380966186523, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4818, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7568922305764411, | |
| "eval_f1": 0.7318195361599812, | |
| "eval_loss": 0.4518365263938904, | |
| "eval_precision": 0.7247646443514644, | |
| "eval_recall": 0.7605019094380796, | |
| "eval_runtime": 5.0705, | |
| "eval_samples_per_second": 78.69, | |
| "eval_steps_per_second": 9.861, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.935502529144287, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4129, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8245614035087719, | |
| "eval_f1": 0.7914426523297491, | |
| "eval_loss": 0.3967439532279968, | |
| "eval_precision": 0.7875574407917992, | |
| "eval_recall": 0.7958719767230406, | |
| "eval_runtime": 5.0798, | |
| "eval_samples_per_second": 78.547, | |
| "eval_steps_per_second": 9.843, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.5287280082702637, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3519, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.849624060150376, | |
| "eval_f1": 0.8176861216035092, | |
| "eval_loss": 0.36260271072387695, | |
| "eval_precision": 0.8193355786895284, | |
| "eval_recall": 0.8161029278050556, | |
| "eval_runtime": 5.0659, | |
| "eval_samples_per_second": 78.761, | |
| "eval_steps_per_second": 9.87, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.9849551916122437, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3191, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8471177944862155, | |
| "eval_f1": 0.8234833375639119, | |
| "eval_loss": 0.3720404803752899, | |
| "eval_precision": 0.8130172220979647, | |
| "eval_recall": 0.8393344244408074, | |
| "eval_runtime": 5.1112, | |
| "eval_samples_per_second": 78.064, | |
| "eval_steps_per_second": 9.782, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.53737735748291, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2977, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8303372434017595, | |
| "eval_loss": 0.3482378423213959, | |
| "eval_precision": 0.8215506437279252, | |
| "eval_recall": 0.8421531187488634, | |
| "eval_runtime": 5.0792, | |
| "eval_samples_per_second": 78.556, | |
| "eval_steps_per_second": 9.844, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.1117504835128784, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2861, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8439163006974941, | |
| "eval_loss": 0.33432629704475403, | |
| "eval_precision": 0.8362546125461254, | |
| "eval_recall": 0.853518821603928, | |
| "eval_runtime": 5.0624, | |
| "eval_samples_per_second": 78.816, | |
| "eval_steps_per_second": 9.877, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.2316811084747314, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2662, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8365204824303285, | |
| "eval_loss": 0.32132402062416077, | |
| "eval_precision": 0.8313636363636363, | |
| "eval_recall": 0.8424713584288053, | |
| "eval_runtime": 5.0679, | |
| "eval_samples_per_second": 78.732, | |
| "eval_steps_per_second": 9.866, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.981521129608154, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2618, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8457993935430168, | |
| "eval_loss": 0.3246487081050873, | |
| "eval_precision": 0.8398540145985401, | |
| "eval_recall": 0.8527914166212038, | |
| "eval_runtime": 5.0499, | |
| "eval_samples_per_second": 79.012, | |
| "eval_steps_per_second": 9.901, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.816589832305908, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2556, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8306964902709584, | |
| "eval_loss": 0.3065463602542877, | |
| "eval_precision": 0.8306964902709584, | |
| "eval_recall": 0.8306964902709584, | |
| "eval_runtime": 5.0629, | |
| "eval_samples_per_second": 78.809, | |
| "eval_steps_per_second": 9.876, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.513470411300659, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2353, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.850481641770208, | |
| "eval_loss": 0.3147808015346527, | |
| "eval_precision": 0.8416070917929654, | |
| "eval_recall": 0.8620658301509365, | |
| "eval_runtime": 5.0812, | |
| "eval_samples_per_second": 78.525, | |
| "eval_steps_per_second": 9.84, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 11.986312866210938, | |
| "learning_rate": 2e-05, | |
| "loss": 0.24, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8524146298159436, | |
| "eval_loss": 0.3097703158855438, | |
| "eval_precision": 0.8451250578971746, | |
| "eval_recall": 0.8613384251682124, | |
| "eval_runtime": 5.1054, | |
| "eval_samples_per_second": 78.153, | |
| "eval_steps_per_second": 9.794, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 1.3587671518325806, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2346, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8771929824561403, | |
| "eval_f1": 0.8514869535493182, | |
| "eval_loss": 0.2989429235458374, | |
| "eval_precision": 0.8523821128305106, | |
| "eval_recall": 0.8506092016730314, | |
| "eval_runtime": 5.1497, | |
| "eval_samples_per_second": 77.481, | |
| "eval_steps_per_second": 9.709, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 11.92502212524414, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2367, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8457993935430168, | |
| "eval_loss": 0.3001403510570526, | |
| "eval_precision": 0.8398540145985401, | |
| "eval_recall": 0.8527914166212038, | |
| "eval_runtime": 5.1188, | |
| "eval_samples_per_second": 77.948, | |
| "eval_steps_per_second": 9.768, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.7083171606063843, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2248, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8498062516145699, | |
| "eval_loss": 0.30395904183387756, | |
| "eval_precision": 0.84200588099631, | |
| "eval_recall": 0.8595653755228223, | |
| "eval_runtime": 5.0513, | |
| "eval_samples_per_second": 78.99, | |
| "eval_steps_per_second": 9.898, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 7.328032493591309, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2174, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8478885630498534, | |
| "eval_loss": 0.30161935091018677, | |
| "eval_precision": 0.8385980403326877, | |
| "eval_recall": 0.8602927805055465, | |
| "eval_runtime": 5.0616, | |
| "eval_samples_per_second": 78.829, | |
| "eval_steps_per_second": 9.878, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 10.082404136657715, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2112, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8391129032258065, | |
| "eval_loss": 0.28866690397262573, | |
| "eval_precision": 0.8345705196182396, | |
| "eval_recall": 0.8442444080741953, | |
| "eval_runtime": 5.0705, | |
| "eval_samples_per_second": 78.691, | |
| "eval_steps_per_second": 9.861, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 5.582000255584717, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2162, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.850481641770208, | |
| "eval_loss": 0.29799067974090576, | |
| "eval_precision": 0.8416070917929654, | |
| "eval_recall": 0.8620658301509365, | |
| "eval_runtime": 5.0708, | |
| "eval_samples_per_second": 78.686, | |
| "eval_steps_per_second": 9.86, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 3.8067166805267334, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2124, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.844327731092437, | |
| "eval_loss": 0.289196640253067, | |
| "eval_precision": 0.8411320530352577, | |
| "eval_recall": 0.8477905073649754, | |
| "eval_runtime": 5.0494, | |
| "eval_samples_per_second": 79.019, | |
| "eval_steps_per_second": 9.902, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 11.847869873046875, | |
| "learning_rate": 0.0, | |
| "loss": 0.2118, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8391129032258065, | |
| "eval_loss": 0.2911982536315918, | |
| "eval_precision": 0.8345705196182396, | |
| "eval_recall": 0.8442444080741953, | |
| "eval_runtime": 5.0549, | |
| "eval_samples_per_second": 78.934, | |
| "eval_steps_per_second": 9.891, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7635661149264000.0, | |
| "train_loss": 0.2866268705149166, | |
| "train_runtime": 1957.4327, | |
| "train_samples_per_second": 37.171, | |
| "train_steps_per_second": 1.247 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7635661149264000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |