| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.3391812865497075, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.023391812865497075, | |
| "eval_loss": 4.397373199462891, | |
| "eval_runtime": 3.0335, | |
| "eval_samples_per_second": 23.735, | |
| "eval_steps_per_second": 2.967, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07017543859649122, | |
| "grad_norm": 3.0865204334259033, | |
| "learning_rate": 1.5e-05, | |
| "loss": 4.4536, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.14035087719298245, | |
| "grad_norm": 3.27763295173645, | |
| "learning_rate": 3e-05, | |
| "loss": 4.304, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "grad_norm": 2.4613616466522217, | |
| "learning_rate": 4.5e-05, | |
| "loss": 3.9912, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "eval_loss": 4.029315948486328, | |
| "eval_runtime": 3.0225, | |
| "eval_samples_per_second": 23.821, | |
| "eval_steps_per_second": 2.978, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2807017543859649, | |
| "grad_norm": 2.587448835372925, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 3.8239, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.3508771929824561, | |
| "grad_norm": 2.6298701763153076, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 3.5308, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.42105263157894735, | |
| "grad_norm": 2.3388280868530273, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 3.5004, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.42105263157894735, | |
| "eval_loss": 3.415943145751953, | |
| "eval_runtime": 3.0398, | |
| "eval_samples_per_second": 23.686, | |
| "eval_steps_per_second": 2.961, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.49122807017543857, | |
| "grad_norm": 2.0435404777526855, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 3.1382, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.5614035087719298, | |
| "grad_norm": 2.0023353099823, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 3.3133, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "grad_norm": 2.8618927001953125, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 3.1938, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "eval_loss": 3.182070732116699, | |
| "eval_runtime": 3.0854, | |
| "eval_samples_per_second": 23.336, | |
| "eval_steps_per_second": 2.917, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7017543859649122, | |
| "grad_norm": 2.293642520904541, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 3.013, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7719298245614035, | |
| "grad_norm": 2.3824496269226074, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 3.1638, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "grad_norm": 2.493609666824341, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 2.8618, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "eval_loss": 3.0471272468566895, | |
| "eval_runtime": 3.0384, | |
| "eval_samples_per_second": 23.697, | |
| "eval_steps_per_second": 2.962, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.9122807017543859, | |
| "grad_norm": 2.103684902191162, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 2.9292, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.9824561403508771, | |
| "grad_norm": 2.324841260910034, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 3.0299, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "grad_norm": 2.2296299934387207, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 3.1419, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "eval_loss": 2.94901180267334, | |
| "eval_runtime": 3.0338, | |
| "eval_samples_per_second": 23.733, | |
| "eval_steps_per_second": 2.967, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.1228070175438596, | |
| "grad_norm": 2.3404273986816406, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 2.8056, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.1929824561403508, | |
| "grad_norm": 2.149913787841797, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 2.7051, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.263157894736842, | |
| "grad_norm": 2.0646862983703613, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 2.7522, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.263157894736842, | |
| "eval_loss": 2.8944597244262695, | |
| "eval_runtime": 3.0449, | |
| "eval_samples_per_second": 23.646, | |
| "eval_steps_per_second": 2.956, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 2.679049491882324, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 2.8691, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.4035087719298245, | |
| "grad_norm": 2.713944911956787, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 2.72, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.4736842105263157, | |
| "grad_norm": 2.994593858718872, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 2.6119, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.4736842105263157, | |
| "eval_loss": 2.8573951721191406, | |
| "eval_runtime": 3.0377, | |
| "eval_samples_per_second": 23.702, | |
| "eval_steps_per_second": 2.963, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.543859649122807, | |
| "grad_norm": 2.503089189529419, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 2.6595, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.6140350877192984, | |
| "grad_norm": 2.928532600402832, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 2.5492, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.6842105263157894, | |
| "grad_norm": 2.7153916358947754, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 2.63, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.6842105263157894, | |
| "eval_loss": 2.8362529277801514, | |
| "eval_runtime": 3.0425, | |
| "eval_samples_per_second": 23.665, | |
| "eval_steps_per_second": 2.958, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.7543859649122808, | |
| "grad_norm": 3.5176753997802734, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 2.9109, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.8245614035087718, | |
| "grad_norm": 3.1322624683380127, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 2.3789, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.8947368421052633, | |
| "grad_norm": 3.4775712490081787, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 3.0337, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.8947368421052633, | |
| "eval_loss": 2.821269989013672, | |
| "eval_runtime": 3.0336, | |
| "eval_samples_per_second": 23.734, | |
| "eval_steps_per_second": 2.967, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.9649122807017543, | |
| "grad_norm": 2.149881601333618, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 2.391, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.0350877192982457, | |
| "grad_norm": 3.047943115234375, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 2.7846, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 3.2259058952331543, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 2.7309, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "eval_loss": 2.8160009384155273, | |
| "eval_runtime": 3.081, | |
| "eval_samples_per_second": 23.369, | |
| "eval_steps_per_second": 2.921, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.175438596491228, | |
| "grad_norm": 2.6105287075042725, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 2.654, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.245614035087719, | |
| "grad_norm": 2.9811198711395264, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 2.3742, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.3157894736842106, | |
| "grad_norm": 3.094942808151245, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 2.7658, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.3157894736842106, | |
| "eval_loss": 2.814375400543213, | |
| "eval_runtime": 3.0526, | |
| "eval_samples_per_second": 23.587, | |
| "eval_steps_per_second": 2.948, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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 | |
| } | |