| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 375, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 1.7216755151748657, | |
| "learning_rate": 5e-05, | |
| "loss": 2.4397, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 1.6662434339523315, | |
| "learning_rate": 9.999782741484788e-05, | |
| "loss": 2.2225, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.1107380390167236, | |
| "learning_rate": 9.904493906342123e-05, | |
| "loss": 1.9769, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 1.3425688743591309, | |
| "learning_rate": 9.639210244594334e-05, | |
| "loss": 1.8957, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.6660774946212769, | |
| "learning_rate": 9.213126762075088e-05, | |
| "loss": 1.9327, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.0810869932174683, | |
| "learning_rate": 8.641011952560371e-05, | |
| "loss": 1.902, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 1.5443850755691528, | |
| "learning_rate": 7.94269590651393e-05, | |
| "loss": 1.8109, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 1.0120549201965332, | |
| "learning_rate": 7.142382979661386e-05, | |
| "loss": 1.7861, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 1.5919784307479858, | |
| "learning_rate": 6.26781284501043e-05, | |
| "loss": 1.8054, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 1.353637933731079, | |
| "learning_rate": 5.3492990071209806e-05, | |
| "loss": 1.7685, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 1.2411092519760132, | |
| "learning_rate": 4.418678104714214e-05, | |
| "loss": 1.8407, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 1.9999362230300903, | |
| "learning_rate": 3.5082064198777e-05, | |
| "loss": 1.7761, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 1.3810490369796753, | |
| "learning_rate": 2.6494418419978482e-05, | |
| "loss": 1.7177, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 1.3220276832580566, | |
| "learning_rate": 1.872150038705015e-05, | |
| "loss": 1.6397, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.6852960586547852, | |
| "learning_rate": 1.203272747076598e-05, | |
| "loss": 1.7622, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 1.4293591976165771, | |
| "learning_rate": 6.659939451910341e-06, | |
| "loss": 1.6719, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 1.658665657043457, | |
| "learning_rate": 2.7893627149161716e-06, | |
| "loss": 1.6602, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 1.8376392126083374, | |
| "learning_rate": 5.551554489528432e-07, | |
| "loss": 1.7036, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 375, | |
| "total_flos": 1382710662660096.0, | |
| "train_loss": 1.8455538279215495, | |
| "train_runtime": 977.2087, | |
| "train_samples_per_second": 3.07, | |
| "train_steps_per_second": 0.384 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 375, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1382710662660096.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |