| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9968, | |
| "eval_steps": 500, | |
| "global_step": 312, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.064, | |
| "grad_norm": 2.304359943215424, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.5192, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 3.367745922034929, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1579, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 0.3120002563540976, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 0.1235, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "grad_norm": 0.34393102439669215, | |
| "learning_rate": 1.9959742939952393e-05, | |
| "loss": 0.1141, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.22846734971092744, | |
| "learning_rate": 1.9796753984232357e-05, | |
| "loss": 0.1154, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 0.22187942259321225, | |
| "learning_rate": 1.9510565162951538e-05, | |
| "loss": 0.1013, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.448, | |
| "grad_norm": 0.1937516417577725, | |
| "learning_rate": 1.9104775466588162e-05, | |
| "loss": 0.0975, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "grad_norm": 0.18776075554849062, | |
| "learning_rate": 1.8584487936018663e-05, | |
| "loss": 0.1003, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 0.16374083531357114, | |
| "learning_rate": 1.795624548881781e-05, | |
| "loss": 0.1011, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.3083074608208316, | |
| "learning_rate": 1.7227948638273918e-05, | |
| "loss": 0.1019, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.704, | |
| "grad_norm": 0.14996630876365036, | |
| "learning_rate": 1.6408756139850243e-05, | |
| "loss": 0.096, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 0.10713257524125, | |
| "learning_rate": 1.5508969814521026e-05, | |
| "loss": 0.0918, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.832, | |
| "grad_norm": 0.16430611401390374, | |
| "learning_rate": 1.4539904997395468e-05, | |
| "loss": 0.0938, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.896, | |
| "grad_norm": 0.13018547178323414, | |
| "learning_rate": 1.3513748240813429e-05, | |
| "loss": 0.0974, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.249248518955414, | |
| "learning_rate": 1.2443404061378941e-05, | |
| "loss": 0.0994, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.9984, | |
| "eval_loss": 0.09072314202785492, | |
| "eval_runtime": 40.5957, | |
| "eval_samples_per_second": 24.633, | |
| "eval_steps_per_second": 1.552, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 1.024, | |
| "grad_norm": 0.13728098111933706, | |
| "learning_rate": 1.1342332658176556e-05, | |
| "loss": 0.088, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.088, | |
| "grad_norm": 0.16275246072291852, | |
| "learning_rate": 1.0224380642958052e-05, | |
| "loss": 0.0947, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.152, | |
| "grad_norm": 0.15001498024621265, | |
| "learning_rate": 9.103606910965666e-06, | |
| "loss": 0.0939, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.216, | |
| "grad_norm": 0.13963224848380856, | |
| "learning_rate": 7.994105842167274e-06, | |
| "loss": 0.0876, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.137367259845856, | |
| "learning_rate": 6.909830056250527e-06, | |
| "loss": 0.0869, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.3439999999999999, | |
| "grad_norm": 0.11133790927990313, | |
| "learning_rate": 5.864414950334796e-06, | |
| "loss": 0.0907, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.408, | |
| "grad_norm": 0.09323151899431767, | |
| "learning_rate": 4.87100722594094e-06, | |
| "loss": 0.0953, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.472, | |
| "grad_norm": 0.15783290675271067, | |
| "learning_rate": 3.942099561591802e-06, | |
| "loss": 0.0911, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.536, | |
| "grad_norm": 0.13377672988354583, | |
| "learning_rate": 3.089373510131354e-06, | |
| "loss": 0.0867, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.1192199124767211, | |
| "learning_rate": 2.323552596419889e-06, | |
| "loss": 0.0871, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.6640000000000001, | |
| "grad_norm": 0.11020679845316991, | |
| "learning_rate": 1.6542674627869738e-06, | |
| "loss": 0.0896, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.728, | |
| "grad_norm": 0.09198353508041567, | |
| "learning_rate": 1.0899347581163222e-06, | |
| "loss": 0.0928, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.792, | |
| "grad_norm": 0.11654532424334699, | |
| "learning_rate": 6.37651293602628e-07, | |
| "loss": 0.0917, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.8559999999999999, | |
| "grad_norm": 0.10356962879053153, | |
| "learning_rate": 3.0310479623313125e-07, | |
| "loss": 0.0847, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.0958617515134905, | |
| "learning_rate": 9.0502382320653e-08, | |
| "loss": 0.0868, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.984, | |
| "grad_norm": 0.08860455805163223, | |
| "learning_rate": 2.5176505749346937e-09, | |
| "loss": 0.0916, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.9968, | |
| "eval_loss": 0.08628984540700912, | |
| "eval_runtime": 40.3884, | |
| "eval_samples_per_second": 24.76, | |
| "eval_steps_per_second": 1.56, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 1.9968, | |
| "step": 312, | |
| "total_flos": 101696396132352.0, | |
| "train_loss": 0.11124454142573552, | |
| "train_runtime": 7872.3532, | |
| "train_samples_per_second": 5.081, | |
| "train_steps_per_second": 0.04 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 312, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 101696396132352.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |