| { | |
| "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": 6.223746299743652, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5424, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7418546365914787, | |
| "eval_f1": 0.6575164379109477, | |
| "eval_loss": 0.4762427806854248, | |
| "eval_precision": 0.6837301587301587, | |
| "eval_recall": 0.647344971813057, | |
| "eval_runtime": 1.7978, | |
| "eval_samples_per_second": 221.935, | |
| "eval_steps_per_second": 27.811, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 4.059021949768066, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4345, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7894736842105263, | |
| "eval_f1": 0.7673663168415792, | |
| "eval_loss": 0.41568055748939514, | |
| "eval_precision": 0.7581367924528302, | |
| "eval_recall": 0.7985542825968357, | |
| "eval_runtime": 1.7989, | |
| "eval_samples_per_second": 221.802, | |
| "eval_steps_per_second": 27.795, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.354827404022217, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.3391, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8546365914786967, | |
| "eval_f1": 0.8180088078011953, | |
| "eval_loss": 0.33879804611206055, | |
| "eval_precision": 0.8323930726843348, | |
| "eval_recall": 0.8071467539552646, | |
| "eval_runtime": 1.8074, | |
| "eval_samples_per_second": 220.763, | |
| "eval_steps_per_second": 27.665, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.747511863708496, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2837, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8621553884711779, | |
| "eval_f1": 0.8333016825553572, | |
| "eval_loss": 0.32792460918426514, | |
| "eval_precision": 0.8341507249908615, | |
| "eval_recall": 0.8324695399163484, | |
| "eval_runtime": 1.804, | |
| "eval_samples_per_second": 221.178, | |
| "eval_steps_per_second": 27.717, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.42053157091140747, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.2761, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8391129032258065, | |
| "eval_loss": 0.31322285532951355, | |
| "eval_precision": 0.8345705196182396, | |
| "eval_recall": 0.8442444080741953, | |
| "eval_runtime": 1.8022, | |
| "eval_samples_per_second": 221.397, | |
| "eval_steps_per_second": 27.744, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.2977254390716553, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2459, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8543795620437956, | |
| "eval_loss": 0.3032587468624115, | |
| "eval_precision": 0.843984962406015, | |
| "eval_recall": 0.868839789052555, | |
| "eval_runtime": 1.8072, | |
| "eval_samples_per_second": 220.78, | |
| "eval_steps_per_second": 27.667, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.8183882236480713, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.2321, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8622085718274466, | |
| "eval_loss": 0.2870577275753021, | |
| "eval_precision": 0.8530168716042322, | |
| "eval_recall": 0.8741589379887251, | |
| "eval_runtime": 1.8055, | |
| "eval_samples_per_second": 220.996, | |
| "eval_steps_per_second": 27.694, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.4162003993988037, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2206, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8822055137844611, | |
| "eval_f1": 0.8560793854229822, | |
| "eval_loss": 0.2634139955043793, | |
| "eval_precision": 0.8609538327526132, | |
| "eval_recall": 0.8516548463356974, | |
| "eval_runtime": 1.8055, | |
| "eval_samples_per_second": 220.985, | |
| "eval_steps_per_second": 27.692, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 5.009228229522705, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.2067, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8703223612108386, | |
| "eval_loss": 0.2633897066116333, | |
| "eval_precision": 0.8694131129742446, | |
| "eval_recall": 0.8712493180578287, | |
| "eval_runtime": 1.8057, | |
| "eval_samples_per_second": 220.963, | |
| "eval_steps_per_second": 27.69, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.9459621906280518, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.192, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8631217838765008, | |
| "eval_loss": 0.2696186900138855, | |
| "eval_precision": 0.8872804935927859, | |
| "eval_recall": 0.8462447717766868, | |
| "eval_runtime": 1.8049, | |
| "eval_samples_per_second": 221.061, | |
| "eval_steps_per_second": 27.702, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.9607306122779846, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.1866, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8808495451466529, | |
| "eval_loss": 0.2752375304698944, | |
| "eval_precision": 0.8691495353421572, | |
| "eval_recall": 0.8972995090016367, | |
| "eval_runtime": 1.8092, | |
| "eval_samples_per_second": 220.542, | |
| "eval_steps_per_second": 27.637, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 4.809903621673584, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1786, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8972431077694235, | |
| "eval_f1": 0.8792560061999484, | |
| "eval_loss": 0.2651856243610382, | |
| "eval_precision": 0.8707622232472325, | |
| "eval_recall": 0.889798145117294, | |
| "eval_runtime": 1.8065, | |
| "eval_samples_per_second": 220.87, | |
| "eval_steps_per_second": 27.678, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.7058310508728027, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.1695, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9072681704260651, | |
| "eval_f1": 0.8867007927797945, | |
| "eval_loss": 0.25362077355384827, | |
| "eval_precision": 0.89198606271777, | |
| "eval_recall": 0.8818876159301692, | |
| "eval_runtime": 1.82, | |
| "eval_samples_per_second": 219.234, | |
| "eval_steps_per_second": 27.473, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 7.428104877471924, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.1664, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8872180451127819, | |
| "eval_f1": 0.8680720368560659, | |
| "eval_loss": 0.2736993730068207, | |
| "eval_precision": 0.8587217615098657, | |
| "eval_recall": 0.8802054919076197, | |
| "eval_runtime": 1.8089, | |
| "eval_samples_per_second": 220.576, | |
| "eval_steps_per_second": 27.641, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.6353508234024048, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1521, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9022556390977443, | |
| "eval_f1": 0.8829621606985718, | |
| "eval_loss": 0.26195329427719116, | |
| "eval_precision": 0.8802419354838709, | |
| "eval_recall": 0.8858428805237315, | |
| "eval_runtime": 1.8126, | |
| "eval_samples_per_second": 220.122, | |
| "eval_steps_per_second": 27.584, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.580483317375183, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1494, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8760914310475572, | |
| "eval_loss": 0.30298247933387756, | |
| "eval_precision": 0.8629851740796268, | |
| "eval_recall": 0.8962538643389707, | |
| "eval_runtime": 1.8105, | |
| "eval_samples_per_second": 220.383, | |
| "eval_steps_per_second": 27.617, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 3.5637781620025635, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.1487, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8733660552828726, | |
| "eval_loss": 0.2702126204967499, | |
| "eval_precision": 0.8650109547970479, | |
| "eval_recall": 0.8837515911983997, | |
| "eval_runtime": 1.8055, | |
| "eval_samples_per_second": 220.99, | |
| "eval_steps_per_second": 27.693, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 3.6467041969299316, | |
| "learning_rate": 5e-06, | |
| "loss": 0.1494, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_f1": 0.8765906680805938, | |
| "eval_loss": 0.2763027548789978, | |
| "eval_precision": 0.8675710594315245, | |
| "eval_recall": 0.888025095471904, | |
| "eval_runtime": 1.8075, | |
| "eval_samples_per_second": 220.753, | |
| "eval_steps_per_second": 27.663, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.27400541305542, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.1334, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8733660552828726, | |
| "eval_loss": 0.28261518478393555, | |
| "eval_precision": 0.8650109547970479, | |
| "eval_recall": 0.8837515911983997, | |
| "eval_runtime": 1.8039, | |
| "eval_samples_per_second": 221.182, | |
| "eval_steps_per_second": 27.717, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.5643185377120972, | |
| "learning_rate": 0.0, | |
| "loss": 0.1325, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8922305764411027, | |
| "eval_f1": 0.8721887408091659, | |
| "eval_loss": 0.27931535243988037, | |
| "eval_precision": 0.866466275659824, | |
| "eval_recall": 0.8787506819421713, | |
| "eval_runtime": 1.804, | |
| "eval_samples_per_second": 221.175, | |
| "eval_steps_per_second": 27.716, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 8444128359504000.0, | |
| "train_loss": 0.2269888150887411, | |
| "train_runtime": 635.9179, | |
| "train_samples_per_second": 114.417, | |
| "train_steps_per_second": 3.837 | |
| } | |
| ], | |
| "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 | |
| } | |