| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.041549807832138776, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00041549807832138777, | |
| "eval_loss": 1.8511379957199097, | |
| "eval_runtime": 172.841, | |
| "eval_samples_per_second": 23.455, | |
| "eval_steps_per_second": 2.933, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0012464942349641633, | |
| "grad_norm": 1.7350784540176392, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.8182, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0024929884699283265, | |
| "grad_norm": 1.777970314025879, | |
| "learning_rate": 3e-05, | |
| "loss": 1.8417, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00373948270489249, | |
| "grad_norm": 1.0666847229003906, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.6831, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00373948270489249, | |
| "eval_loss": 1.645044207572937, | |
| "eval_runtime": 173.3911, | |
| "eval_samples_per_second": 23.381, | |
| "eval_steps_per_second": 2.924, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004985976939856653, | |
| "grad_norm": 1.074853777885437, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 1.638, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0062324711748208165, | |
| "grad_norm": 1.2451493740081787, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 1.4395, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00747896540978498, | |
| "grad_norm": 0.8546340465545654, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 1.4021, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00747896540978498, | |
| "eval_loss": 1.383123517036438, | |
| "eval_runtime": 173.2897, | |
| "eval_samples_per_second": 23.394, | |
| "eval_steps_per_second": 2.926, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.008725459644749143, | |
| "grad_norm": 0.8779253959655762, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.3657, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.009971953879713306, | |
| "grad_norm": 0.9482561945915222, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.2238, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01121844811467747, | |
| "grad_norm": 0.7827633023262024, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 1.2321, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01121844811467747, | |
| "eval_loss": 1.3266533613204956, | |
| "eval_runtime": 173.524, | |
| "eval_samples_per_second": 23.363, | |
| "eval_steps_per_second": 2.922, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.012464942349641633, | |
| "grad_norm": 0.8669916987419128, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 1.2845, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.013711436584605797, | |
| "grad_norm": 0.8390219211578369, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 1.3824, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.01495793081956996, | |
| "grad_norm": 0.7428231835365295, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 1.2802, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01495793081956996, | |
| "eval_loss": 1.293402075767517, | |
| "eval_runtime": 173.4725, | |
| "eval_samples_per_second": 23.37, | |
| "eval_steps_per_second": 2.923, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.016204425054534122, | |
| "grad_norm": 0.7994734048843384, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 1.3122, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.017450919289498285, | |
| "grad_norm": 0.7998748421669006, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 1.3337, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.01869741352446245, | |
| "grad_norm": 0.7622133493423462, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 1.3281, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.01869741352446245, | |
| "eval_loss": 1.2758731842041016, | |
| "eval_runtime": 173.596, | |
| "eval_samples_per_second": 23.353, | |
| "eval_steps_per_second": 2.921, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.019943907759426612, | |
| "grad_norm": 0.7918552160263062, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 1.2846, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.021190401994390776, | |
| "grad_norm": 0.7809180617332458, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 1.275, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.02243689622935494, | |
| "grad_norm": 0.6568572521209717, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 1.2913, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.02243689622935494, | |
| "eval_loss": 1.263215184211731, | |
| "eval_runtime": 173.5147, | |
| "eval_samples_per_second": 23.364, | |
| "eval_steps_per_second": 2.922, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.023683390464319103, | |
| "grad_norm": 0.6685330271720886, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 1.1958, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.024929884699283266, | |
| "grad_norm": 0.7554190158843994, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 1.1446, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.02617637893424743, | |
| "grad_norm": 0.7289713025093079, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 1.2659, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.02617637893424743, | |
| "eval_loss": 1.2556233406066895, | |
| "eval_runtime": 173.4263, | |
| "eval_samples_per_second": 23.376, | |
| "eval_steps_per_second": 2.923, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.027422873169211593, | |
| "grad_norm": 0.7362277507781982, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 1.2666, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.028669367404175757, | |
| "grad_norm": 0.7522065043449402, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 1.2857, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.02991586163913992, | |
| "grad_norm": 0.7410892844200134, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 1.1732, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.02991586163913992, | |
| "eval_loss": 1.250916600227356, | |
| "eval_runtime": 173.421, | |
| "eval_samples_per_second": 23.377, | |
| "eval_steps_per_second": 2.924, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.031162355874104083, | |
| "grad_norm": 0.8000510334968567, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 1.2934, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.032408850109068243, | |
| "grad_norm": 0.8692224621772766, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 1.37, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.03365534434403241, | |
| "grad_norm": 0.8634310364723206, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 1.2589, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.03365534434403241, | |
| "eval_loss": 1.2482656240463257, | |
| "eval_runtime": 173.4527, | |
| "eval_samples_per_second": 23.372, | |
| "eval_steps_per_second": 2.923, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.03490183857899657, | |
| "grad_norm": 0.7589036822319031, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 1.1979, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.03614833281396074, | |
| "grad_norm": 0.6934281587600708, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 1.3394, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.0373948270489249, | |
| "grad_norm": 0.775426983833313, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 1.2419, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0373948270489249, | |
| "eval_loss": 1.2471787929534912, | |
| "eval_runtime": 173.3007, | |
| "eval_samples_per_second": 23.393, | |
| "eval_steps_per_second": 2.926, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.038641321283889064, | |
| "grad_norm": 0.6684338450431824, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 1.167, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.039887815518853224, | |
| "grad_norm": 0.700363278388977, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 1.2644, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.04113430975381739, | |
| "grad_norm": 0.7705054879188538, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 1.2685, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.04113430975381739, | |
| "eval_loss": 1.2470118999481201, | |
| "eval_runtime": 173.4404, | |
| "eval_samples_per_second": 23.374, | |
| "eval_steps_per_second": 2.923, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.041299127907123e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |