| { | |
| "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": 0.08, | |
| "grad_norm": 1.4892003536224365, | |
| "learning_rate": 5.263157894736842e-06, | |
| "loss": 0.4583, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 2.1322124004364014, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 0.4483, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 1.2268545627593994, | |
| "learning_rate": 1.578947368421053e-05, | |
| "loss": 0.4244, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.5979569554328918, | |
| "learning_rate": 1.9998261969639324e-05, | |
| "loss": 0.4404, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.507793664932251, | |
| "learning_rate": 1.9937494319239112e-05, | |
| "loss": 0.4429, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.45057329535484314, | |
| "learning_rate": 1.979042835741503e-05, | |
| "loss": 0.4641, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.5177234411239624, | |
| "learning_rate": 1.9558341221417744e-05, | |
| "loss": 0.4256, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.45696133375167847, | |
| "learning_rate": 1.9243248381877605e-05, | |
| "loss": 0.4211, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.4730648100376129, | |
| "learning_rate": 1.8847886140232438e-05, | |
| "loss": 0.4311, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.4379076361656189, | |
| "learning_rate": 1.8375687866379988e-05, | |
| "loss": 0.4364, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.49525463581085205, | |
| "learning_rate": 1.7830754182909985e-05, | |
| "loss": 0.4272, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.5524850487709045, | |
| "learning_rate": 1.721781735483921e-05, | |
| "loss": 0.4294, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.35776063799858093, | |
| "learning_rate": 1.654220019409317e-05, | |
| "loss": 0.3735, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.45718610286712646, | |
| "learning_rate": 1.580976983561235e-05, | |
| "loss": 0.3037, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.5565239787101746, | |
| "learning_rate": 1.5026886786496624e-05, | |
| "loss": 0.3217, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.5376655459403992, | |
| "learning_rate": 1.4200349690650654e-05, | |
| "loss": 0.2992, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 0.5882963538169861, | |
| "learning_rate": 1.3337336288600297e-05, | |
| "loss": 0.3231, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.47769567370414734, | |
| "learning_rate": 1.24453410851916e-05, | |
| "loss": 0.3168, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.5084702968597412, | |
| "learning_rate": 1.1532110266473026e-05, | |
| "loss": 0.3143, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.400981068611145, | |
| "learning_rate": 1.0605574430949983e-05, | |
| "loss": 0.314, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 0.998202919960022, | |
| "learning_rate": 9.673779719380967e-06, | |
| "loss": 0.3328, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.47775644063949585, | |
| "learning_rate": 8.744817941191862e-06, | |
| "loss": 0.3195, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 0.4177009165287018, | |
| "learning_rate": 7.826756304298428e-06, | |
| "loss": 0.3037, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.382407546043396, | |
| "learning_rate": 6.92756735857107e-06, | |
| "loss": 0.3294, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.3983019292354584, | |
| "learning_rate": 6.0550597613206205e-06, | |
| "loss": 0.3075, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.4541538655757904, | |
| "learning_rate": 5.216810466045448e-06, | |
| "loss": 0.2471, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 0.36751970648765564, | |
| "learning_rate": 4.420098923320378e-06, | |
| "loss": 0.2303, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.5258793830871582, | |
| "learning_rate": 3.671843865234238e-06, | |
| "loss": 0.2229, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.4654676020145416, | |
| "learning_rate": 2.978543222347076e-06, | |
| "loss": 0.2207, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.4600491523742676, | |
| "learning_rate": 2.346217694934847e-06, | |
| "loss": 0.2331, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.34612563252449036, | |
| "learning_rate": 1.7803584685552877e-06, | |
| "loss": 0.2288, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.44249340891838074, | |
| "learning_rate": 1.2858795279787517e-06, | |
| "loss": 0.2238, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 0.45505207777023315, | |
| "learning_rate": 8.670749835951964e-07, | |
| "loss": 0.2432, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 0.37717291712760925, | |
| "learning_rate": 5.275817808796013e-07, | |
| "loss": 0.2273, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.45334893465042114, | |
| "learning_rate": 2.703481167509281e-07, | |
| "loss": 0.2203, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.4163905382156372, | |
| "learning_rate": 9.760783710056176e-08, | |
| "loss": 0.2377, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 0.3816494047641754, | |
| "learning_rate": 1.0861037824896337e-08, | |
| "loss": 0.2295, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 375, | |
| "total_flos": 74004831797248.0, | |
| "train_loss": 0.32720418787002564, | |
| "train_runtime": 6412.3432, | |
| "train_samples_per_second": 0.468, | |
| "train_steps_per_second": 0.058 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 375, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 74004831797248.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |