| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.008510276158461342, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.510276158461342e-05, | |
| "eval_loss": 1.4265342950820923, | |
| "eval_runtime": 1420.4288, | |
| "eval_samples_per_second": 13.933, | |
| "eval_steps_per_second": 1.742, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002553082847538403, | |
| "grad_norm": 8.264724731445312, | |
| "learning_rate": 1.5e-05, | |
| "loss": 5.6967, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0005106165695076806, | |
| "grad_norm": 6.364163875579834, | |
| "learning_rate": 3e-05, | |
| "loss": 5.4301, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0007659248542615208, | |
| "grad_norm": 4.069925785064697, | |
| "learning_rate": 4.5e-05, | |
| "loss": 5.2099, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0007659248542615208, | |
| "eval_loss": 1.2636969089508057, | |
| "eval_runtime": 1429.8254, | |
| "eval_samples_per_second": 13.842, | |
| "eval_steps_per_second": 1.73, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0010212331390153612, | |
| "grad_norm": 4.338223934173584, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 5.0681, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0012765414237692013, | |
| "grad_norm": 2.9977951049804688, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 4.8176, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0015318497085230415, | |
| "grad_norm": 3.106626272201538, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 4.7295, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0015318497085230415, | |
| "eval_loss": 1.1791139841079712, | |
| "eval_runtime": 1428.3193, | |
| "eval_samples_per_second": 13.856, | |
| "eval_steps_per_second": 1.732, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.001787157993276882, | |
| "grad_norm": 2.7620089054107666, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 4.693, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0020424662780307223, | |
| "grad_norm": 2.6547420024871826, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 4.5969, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0022977745627845623, | |
| "grad_norm": 2.783263921737671, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 4.6329, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0022977745627845623, | |
| "eval_loss": 1.1524362564086914, | |
| "eval_runtime": 1428.632, | |
| "eval_samples_per_second": 13.853, | |
| "eval_steps_per_second": 1.732, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0025530828475384027, | |
| "grad_norm": 2.664246082305908, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 4.5762, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.002808391132292243, | |
| "grad_norm": 2.6894233226776123, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 4.6249, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.003063699417046083, | |
| "grad_norm": 2.957350730895996, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 4.5452, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.003063699417046083, | |
| "eval_loss": 1.1378833055496216, | |
| "eval_runtime": 1430.8088, | |
| "eval_samples_per_second": 13.832, | |
| "eval_steps_per_second": 1.729, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0033190077017999234, | |
| "grad_norm": 2.741870164871216, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 4.6275, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.003574315986553764, | |
| "grad_norm": 2.8776938915252686, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 4.5287, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0038296242713076038, | |
| "grad_norm": 2.7638866901397705, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 4.4224, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0038296242713076038, | |
| "eval_loss": 1.1293689012527466, | |
| "eval_runtime": 1429.6849, | |
| "eval_samples_per_second": 13.843, | |
| "eval_steps_per_second": 1.73, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004084932556061445, | |
| "grad_norm": 2.8020057678222656, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 4.3762, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.004340240840815284, | |
| "grad_norm": 2.6722795963287354, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 4.5613, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.0045955491255691245, | |
| "grad_norm": 2.5772705078125, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 4.5655, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0045955491255691245, | |
| "eval_loss": 1.1215273141860962, | |
| "eval_runtime": 1429.4909, | |
| "eval_samples_per_second": 13.845, | |
| "eval_steps_per_second": 1.731, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.004850857410322965, | |
| "grad_norm": 2.7015180587768555, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 4.4799, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.005106165695076805, | |
| "grad_norm": 2.699486017227173, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 4.4337, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.005361473979830646, | |
| "grad_norm": 2.7420105934143066, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 4.4129, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.005361473979830646, | |
| "eval_loss": 1.1171752214431763, | |
| "eval_runtime": 1430.265, | |
| "eval_samples_per_second": 13.837, | |
| "eval_steps_per_second": 1.73, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.005616782264584486, | |
| "grad_norm": 2.7122933864593506, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 4.4688, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.005872090549338326, | |
| "grad_norm": 2.925189971923828, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 4.428, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.006127398834092166, | |
| "grad_norm": 2.7338027954101562, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 4.3209, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.006127398834092166, | |
| "eval_loss": 1.1138705015182495, | |
| "eval_runtime": 1429.7869, | |
| "eval_samples_per_second": 13.842, | |
| "eval_steps_per_second": 1.73, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0063827071188460064, | |
| "grad_norm": 2.891840696334839, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 4.4975, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.006638015403599847, | |
| "grad_norm": 2.6543121337890625, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 4.3666, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.006893323688353687, | |
| "grad_norm": 2.696539878845215, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 4.3767, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.006893323688353687, | |
| "eval_loss": 1.1115138530731201, | |
| "eval_runtime": 1428.6614, | |
| "eval_samples_per_second": 13.853, | |
| "eval_steps_per_second": 1.732, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.007148631973107528, | |
| "grad_norm": 2.8411285877227783, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 4.5361, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.007403940257861368, | |
| "grad_norm": 2.6249213218688965, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 4.4003, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.0076592485426152076, | |
| "grad_norm": 2.738430976867676, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 4.3995, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.0076592485426152076, | |
| "eval_loss": 1.1105740070343018, | |
| "eval_runtime": 1429.5163, | |
| "eval_samples_per_second": 13.845, | |
| "eval_steps_per_second": 1.731, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.007914556827369048, | |
| "grad_norm": 2.819993734359741, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 4.5155, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.00816986511212289, | |
| "grad_norm": 2.8105878829956055, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 4.5114, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.008425173396876729, | |
| "grad_norm": 2.747713327407837, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 4.4013, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.008425173396876729, | |
| "eval_loss": 1.1103698015213013, | |
| "eval_runtime": 1428.4413, | |
| "eval_samples_per_second": 13.855, | |
| "eval_steps_per_second": 1.732, | |
| "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": 1.406260607975424e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |