| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0028138629648736106, | |
| "eval_steps": 4, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 9.379543216245368e-05, | |
| "grad_norm": 1.8281327486038208, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.6653, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 9.379543216245368e-05, | |
| "eval_loss": 2.1829569339752197, | |
| "eval_runtime": 566.6389, | |
| "eval_samples_per_second": 7.924, | |
| "eval_steps_per_second": 1.982, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00018759086432490737, | |
| "grad_norm": 2.9964587688446045, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.4715, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0002813862964873611, | |
| "grad_norm": 1.8189451694488525, | |
| "learning_rate": 6e-06, | |
| "loss": 1.9373, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00037518172864981473, | |
| "grad_norm": 3.3282887935638428, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.6139, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00037518172864981473, | |
| "eval_loss": 2.182229995727539, | |
| "eval_runtime": 567.0595, | |
| "eval_samples_per_second": 7.918, | |
| "eval_steps_per_second": 1.98, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00046897716081226843, | |
| "grad_norm": 1.7374569177627563, | |
| "learning_rate": 1e-05, | |
| "loss": 1.5677, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0005627725929747222, | |
| "grad_norm": 3.907545804977417, | |
| "learning_rate": 9.960573506572391e-06, | |
| "loss": 3.4244, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0006565680251371758, | |
| "grad_norm": 2.272611379623413, | |
| "learning_rate": 9.842915805643156e-06, | |
| "loss": 1.8311, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0007503634572996295, | |
| "grad_norm": 3.4874234199523926, | |
| "learning_rate": 9.648882429441258e-06, | |
| "loss": 2.7307, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0007503634572996295, | |
| "eval_loss": 2.1755871772766113, | |
| "eval_runtime": 567.3081, | |
| "eval_samples_per_second": 7.915, | |
| "eval_steps_per_second": 1.98, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0008441588894620832, | |
| "grad_norm": 2.1225948333740234, | |
| "learning_rate": 9.381533400219319e-06, | |
| "loss": 1.8554, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0009379543216245369, | |
| "grad_norm": 1.8341472148895264, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 2.0655, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0010317497537869905, | |
| "grad_norm": 2.316349983215332, | |
| "learning_rate": 8.644843137107058e-06, | |
| "loss": 1.8317, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0011255451859494444, | |
| "grad_norm": 3.063673734664917, | |
| "learning_rate": 8.18711994874345e-06, | |
| "loss": 2.5746, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0011255451859494444, | |
| "eval_loss": 2.1628923416137695, | |
| "eval_runtime": 567.5155, | |
| "eval_samples_per_second": 7.912, | |
| "eval_steps_per_second": 1.979, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.001219340618111898, | |
| "grad_norm": 1.8529959917068481, | |
| "learning_rate": 7.679133974894984e-06, | |
| "loss": 1.2502, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0013131360502743516, | |
| "grad_norm": 2.0128021240234375, | |
| "learning_rate": 7.128896457825364e-06, | |
| "loss": 1.7283, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0014069314824368053, | |
| "grad_norm": 2.0994114875793457, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 1.8423, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.001500726914599259, | |
| "grad_norm": 1.1801517009735107, | |
| "learning_rate": 5.936906572928625e-06, | |
| "loss": 0.6602, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.001500726914599259, | |
| "eval_loss": 2.146209478378296, | |
| "eval_runtime": 567.8634, | |
| "eval_samples_per_second": 7.907, | |
| "eval_steps_per_second": 1.978, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0015945223467617128, | |
| "grad_norm": 1.2016880512237549, | |
| "learning_rate": 5.3139525976465675e-06, | |
| "loss": 1.6955, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0016883177789241664, | |
| "grad_norm": 2.6693007946014404, | |
| "learning_rate": 4.686047402353433e-06, | |
| "loss": 2.4726, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00178211321108662, | |
| "grad_norm": 2.235560417175293, | |
| "learning_rate": 4.063093427071376e-06, | |
| "loss": 1.8914, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0018759086432490737, | |
| "grad_norm": 1.5127105712890625, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 1.6971, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0018759086432490737, | |
| "eval_loss": 2.131486177444458, | |
| "eval_runtime": 567.9739, | |
| "eval_samples_per_second": 7.905, | |
| "eval_steps_per_second": 1.977, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0019697040754115274, | |
| "grad_norm": 2.2492661476135254, | |
| "learning_rate": 2.871103542174637e-06, | |
| "loss": 1.9397, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.002063499507573981, | |
| "grad_norm": 2.5874478816986084, | |
| "learning_rate": 2.320866025105016e-06, | |
| "loss": 2.4379, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0021572949397364346, | |
| "grad_norm": 2.168783187866211, | |
| "learning_rate": 1.8128800512565514e-06, | |
| "loss": 2.5954, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0022510903718988887, | |
| "grad_norm": 1.9641475677490234, | |
| "learning_rate": 1.3551568628929434e-06, | |
| "loss": 2.2953, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0022510903718988887, | |
| "eval_loss": 2.123440980911255, | |
| "eval_runtime": 568.41, | |
| "eval_samples_per_second": 7.899, | |
| "eval_steps_per_second": 1.976, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0023448858040613424, | |
| "grad_norm": 2.4187934398651123, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 1.7173, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.002438681236223796, | |
| "grad_norm": 1.4847743511199951, | |
| "learning_rate": 6.184665997806832e-07, | |
| "loss": 0.83, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0025324766683862496, | |
| "grad_norm": 3.2388100624084473, | |
| "learning_rate": 3.511175705587433e-07, | |
| "loss": 2.4771, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0026262721005487033, | |
| "grad_norm": 2.044548273086548, | |
| "learning_rate": 1.5708419435684463e-07, | |
| "loss": 1.4278, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0026262721005487033, | |
| "eval_loss": 2.120626211166382, | |
| "eval_runtime": 567.8072, | |
| "eval_samples_per_second": 7.908, | |
| "eval_steps_per_second": 1.978, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.002720067532711157, | |
| "grad_norm": 2.0032973289489746, | |
| "learning_rate": 3.9426493427611177e-08, | |
| "loss": 2.2627, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0028138629648736106, | |
| "grad_norm": 2.9637610912323, | |
| "learning_rate": 0.0, | |
| "loss": 2.4295, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 20, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3539442924978176e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |