| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 132, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 9989.583355258865, | |
| "learning_rate": 1.4285714285714286e-06, | |
| "loss": 29.4449, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 707.0577616119402, | |
| "learning_rate": 7.1428571428571436e-06, | |
| "loss": 22.2193, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 375.6314604650611, | |
| "learning_rate": 1.4285714285714287e-05, | |
| "loss": 15.1108, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 289.1773570401075, | |
| "learning_rate": 1.999645611123453e-05, | |
| "loss": 12.8322, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 100.15622758708129, | |
| "learning_rate": 1.9872683547213446e-05, | |
| "loss": 11.4674, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 304.81520026489596, | |
| "learning_rate": 1.9574220383620054e-05, | |
| "loss": 10.0101, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 371.0345686284941, | |
| "learning_rate": 1.9106347728549134e-05, | |
| "loss": 6.2458, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 81.2827670282697, | |
| "learning_rate": 1.8477344278896708e-05, | |
| "loss": 3.0864, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 47.894252021447926, | |
| "learning_rate": 1.7698339834299064e-05, | |
| "loss": 2.8755, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 53.028055369579555, | |
| "learning_rate": 1.6783118362696162e-05, | |
| "loss": 2.6268, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 50.2509028556341, | |
| "learning_rate": 1.5747874102144073e-05, | |
| "loss": 2.3685, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 20.877545879705412, | |
| "learning_rate": 1.461092501449326e-05, | |
| "loss": 2.2533, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 36.31764261618316, | |
| "learning_rate": 1.3392388661180303e-05, | |
| "loss": 2.1889, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 29.627487525974026, | |
| "learning_rate": 1.2113826236296245e-05, | |
| "loss": 2.0759, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 19.76492979476836, | |
| "learning_rate": 1.0797861055530832e-05, | |
| "loss": 2.0685, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 8.65646635592006, | |
| "learning_rate": 9.467778251578217e-06, | |
| "loss": 2.0652, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 21.898667968620007, | |
| "learning_rate": 8.147112759128859e-06, | |
| "loss": 2.0181, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 7.552721457027234, | |
| "learning_rate": 6.859232879780515e-06, | |
| "loss": 2.0121, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 26.791885422498936, | |
| "learning_rate": 5.626926795411447e-06, | |
| "loss": 1.9533, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 13.308383024928718, | |
| "learning_rate": 4.4719993463880695e-06, | |
| "loss": 1.8816, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "grad_norm": 10.35688603485077, | |
| "learning_rate": 3.414886209349615e-06, | |
| "loss": 1.8907, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 14.11012256128135, | |
| "learning_rate": 2.4742923014386154e-06, | |
| "loss": 1.8785, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 8.840068155810476, | |
| "learning_rate": 1.6668608091748495e-06, | |
| "loss": 1.896, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 10.535980554072063, | |
| "learning_rate": 1.0068786982878087e-06, | |
| "loss": 1.8747, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 7.2944918703673345, | |
| "learning_rate": 5.060239153161872e-07, | |
| "loss": 1.8646, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 6.536602991990556, | |
| "learning_rate": 1.731587540747903e-07, | |
| "loss": 1.8607, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 8.045080822985964, | |
| "learning_rate": 1.4173043232380557e-08, | |
| "loss": 1.8633, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 132, | |
| "total_flos": 9088687669248.0, | |
| "train_loss": 4.646637526425448, | |
| "train_runtime": 252.9729, | |
| "train_samples_per_second": 16.614, | |
| "train_steps_per_second": 0.522 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 132, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 9088687669248.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |