| { | |
| "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": 3.324044704437256, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.5577, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7192982456140351, | |
| "eval_f1": 0.6738102189781022, | |
| "eval_loss": 0.5371509194374084, | |
| "eval_precision": 0.6691729323308271, | |
| "eval_recall": 0.6813966175668303, | |
| "eval_runtime": 1.7143, | |
| "eval_samples_per_second": 232.75, | |
| "eval_steps_per_second": 29.167, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.0211801528930664, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5012, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7669172932330827, | |
| "eval_f1": 0.6651023005207534, | |
| "eval_loss": 0.47705554962158203, | |
| "eval_precision": 0.7369819086819183, | |
| "eval_recall": 0.6500727404982725, | |
| "eval_runtime": 1.7201, | |
| "eval_samples_per_second": 231.96, | |
| "eval_steps_per_second": 29.068, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 5.67899227142334, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.4626, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8070175438596491, | |
| "eval_f1": 0.7504244482173175, | |
| "eval_loss": 0.4250223934650421, | |
| "eval_precision": 0.7756337629230987, | |
| "eval_recall": 0.7359519912711402, | |
| "eval_runtime": 1.7239, | |
| "eval_samples_per_second": 231.449, | |
| "eval_steps_per_second": 29.004, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.776041030883789, | |
| "learning_rate": 4e-05, | |
| "loss": 0.4055, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8345864661654135, | |
| "eval_f1": 0.8024159663865547, | |
| "eval_loss": 0.3896389603614807, | |
| "eval_precision": 0.7996165051429931, | |
| "eval_recall": 0.8054646299327151, | |
| "eval_runtime": 1.7227, | |
| "eval_samples_per_second": 231.613, | |
| "eval_steps_per_second": 29.024, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.4831624031066895, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.3709, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8295739348370927, | |
| "eval_f1": 0.7912280701754386, | |
| "eval_loss": 0.3578239679336548, | |
| "eval_precision": 0.7961242492492493, | |
| "eval_recall": 0.7869158028732497, | |
| "eval_runtime": 1.7277, | |
| "eval_samples_per_second": 230.944, | |
| "eval_steps_per_second": 28.94, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.8411459922790527, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.3385, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8370927318295739, | |
| "eval_f1": 0.8085765951950401, | |
| "eval_loss": 0.3523370623588562, | |
| "eval_precision": 0.8017470018450185, | |
| "eval_recall": 0.817239498090562, | |
| "eval_runtime": 1.7391, | |
| "eval_samples_per_second": 229.429, | |
| "eval_steps_per_second": 28.75, | |
| "step": 732 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.7591593265533447, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.3276, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8521303258145363, | |
| "eval_f1": 0.816408656658218, | |
| "eval_loss": 0.33066093921661377, | |
| "eval_precision": 0.8270654903728508, | |
| "eval_recall": 0.8078741589379888, | |
| "eval_runtime": 1.7333, | |
| "eval_samples_per_second": 230.193, | |
| "eval_steps_per_second": 28.846, | |
| "step": 854 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.4650044441223145, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3133, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_f1": 0.8196102381877741, | |
| "eval_loss": 0.32557404041290283, | |
| "eval_precision": 0.8381270903010034, | |
| "eval_recall": 0.8064193489725404, | |
| "eval_runtime": 1.7351, | |
| "eval_samples_per_second": 229.956, | |
| "eval_steps_per_second": 28.816, | |
| "step": 976 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 6.412709712982178, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.3039, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8286082474226804, | |
| "eval_loss": 0.32820799946784973, | |
| "eval_precision": 0.8490909090909091, | |
| "eval_recall": 0.8142389525368249, | |
| "eval_runtime": 1.742, | |
| "eval_samples_per_second": 229.046, | |
| "eval_steps_per_second": 28.703, | |
| "step": 1098 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.903890609741211, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2831, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8596491228070176, | |
| "eval_f1": 0.8298403801632752, | |
| "eval_loss": 0.3141604959964752, | |
| "eval_precision": 0.8315523576240049, | |
| "eval_recall": 0.8281960356428442, | |
| "eval_runtime": 1.7387, | |
| "eval_samples_per_second": 229.485, | |
| "eval_steps_per_second": 28.758, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 2.507814884185791, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.2798, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.87468671679198, | |
| "eval_f1": 0.8464912280701754, | |
| "eval_loss": 0.3033808767795563, | |
| "eval_precision": 0.8522897897897899, | |
| "eval_recall": 0.8413347881432988, | |
| "eval_runtime": 1.7365, | |
| "eval_samples_per_second": 229.779, | |
| "eval_steps_per_second": 28.794, | |
| "step": 1342 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 3.706669330596924, | |
| "learning_rate": 2e-05, | |
| "loss": 0.269, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8341632880321839, | |
| "eval_loss": 0.30022966861724854, | |
| "eval_precision": 0.8479139504563233, | |
| "eval_recall": 0.8235133660665576, | |
| "eval_runtime": 1.7385, | |
| "eval_samples_per_second": 229.511, | |
| "eval_steps_per_second": 28.761, | |
| "step": 1464 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 5.18087100982666, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.2699, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.842789598108747, | |
| "eval_loss": 0.29726317524909973, | |
| "eval_precision": 0.842789598108747, | |
| "eval_recall": 0.842789598108747, | |
| "eval_runtime": 1.7395, | |
| "eval_samples_per_second": 229.378, | |
| "eval_steps_per_second": 28.744, | |
| "step": 1586 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 3.581437110900879, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2657, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8469505178365937, | |
| "eval_loss": 0.29847633838653564, | |
| "eval_precision": 0.844489247311828, | |
| "eval_recall": 0.8495635570103655, | |
| "eval_runtime": 1.7412, | |
| "eval_samples_per_second": 229.157, | |
| "eval_steps_per_second": 28.716, | |
| "step": 1708 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 3.276730537414551, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2537, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8377065410088949, | |
| "eval_loss": 0.28864598274230957, | |
| "eval_precision": 0.8423344947735192, | |
| "eval_recall": 0.8335151845790143, | |
| "eval_runtime": 1.7378, | |
| "eval_samples_per_second": 229.607, | |
| "eval_steps_per_second": 28.773, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 2.79162335395813, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2529, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8646616541353384, | |
| "eval_f1": 0.8350734843845683, | |
| "eval_loss": 0.2877609431743622, | |
| "eval_precision": 0.8386812302741506, | |
| "eval_recall": 0.8317421349336243, | |
| "eval_runtime": 1.7331, | |
| "eval_samples_per_second": 230.23, | |
| "eval_steps_per_second": 28.851, | |
| "step": 1952 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 6.254048824310303, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.2565, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8721804511278195, | |
| "eval_f1": 0.8454251965513313, | |
| "eval_loss": 0.28774765133857727, | |
| "eval_precision": 0.8463049835506276, | |
| "eval_recall": 0.8445626477541371, | |
| "eval_runtime": 1.7362, | |
| "eval_samples_per_second": 229.817, | |
| "eval_steps_per_second": 28.799, | |
| "step": 2074 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 2.5236241817474365, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2514, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8385441718775052, | |
| "eval_loss": 0.28566497564315796, | |
| "eval_precision": 0.8412280701754387, | |
| "eval_recall": 0.8360156392071285, | |
| "eval_runtime": 1.7372, | |
| "eval_samples_per_second": 229.685, | |
| "eval_steps_per_second": 28.783, | |
| "step": 2196 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 4.119316577911377, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.2517, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8696741854636592, | |
| "eval_f1": 0.8419946387230413, | |
| "eval_loss": 0.28444331884384155, | |
| "eval_precision": 0.8437691365584814, | |
| "eval_recall": 0.8402891434806329, | |
| "eval_runtime": 1.7434, | |
| "eval_samples_per_second": 228.858, | |
| "eval_steps_per_second": 28.679, | |
| "step": 2318 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.4626479148864746, | |
| "learning_rate": 0.0, | |
| "loss": 0.2512, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8671679197994987, | |
| "eval_f1": 0.8377065410088949, | |
| "eval_loss": 0.2846580445766449, | |
| "eval_precision": 0.8423344947735192, | |
| "eval_recall": 0.8335151845790143, | |
| "eval_runtime": 1.7409, | |
| "eval_samples_per_second": 229.195, | |
| "eval_steps_per_second": 28.721, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2440, | |
| "total_flos": 7624554283800000.0, | |
| "train_loss": 0.32331084657887943, | |
| "train_runtime": 627.3488, | |
| "train_samples_per_second": 116.203, | |
| "train_steps_per_second": 3.889 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 7624554283800000.0, | |
| "train_batch_size": 30, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |