| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.015786565632646618, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00031573131265293233, | |
| "grad_norm": 0.6405626535415649, | |
| "learning_rate": 2e-05, | |
| "loss": 6.5486, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00031573131265293233, | |
| "eval_loss": 1.742234706878662, | |
| "eval_runtime": 74.1238, | |
| "eval_samples_per_second": 17.997, | |
| "eval_steps_per_second": 8.998, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0006314626253058647, | |
| "grad_norm": 0.500285267829895, | |
| "learning_rate": 4e-05, | |
| "loss": 6.1622, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.000947193937958797, | |
| "grad_norm": 0.6612991690635681, | |
| "learning_rate": 6e-05, | |
| "loss": 7.4267, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0012629252506117293, | |
| "grad_norm": 0.6658630967140198, | |
| "learning_rate": 8e-05, | |
| "loss": 6.6306, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0015786565632646618, | |
| "grad_norm": 0.7246699333190918, | |
| "learning_rate": 0.0001, | |
| "loss": 6.792, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.001894387875917594, | |
| "grad_norm": 0.6887425780296326, | |
| "learning_rate": 0.00012, | |
| "loss": 6.5042, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0022101191885705264, | |
| "grad_norm": 1.1644823551177979, | |
| "learning_rate": 0.00014, | |
| "loss": 6.5081, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0025258505012234586, | |
| "grad_norm": 1.0138639211654663, | |
| "learning_rate": 0.00016, | |
| "loss": 6.8784, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0028415818138763913, | |
| "grad_norm": 1.9288033246994019, | |
| "learning_rate": 0.00018, | |
| "loss": 7.035, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0031573131265293236, | |
| "grad_norm": 2.2389585971832275, | |
| "learning_rate": 0.0002, | |
| "loss": 6.8514, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.003473044439182256, | |
| "grad_norm": 1.7787041664123535, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 5.8545, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.003788775751835188, | |
| "grad_norm": 1.6030915975570679, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 5.7655, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.004104507064488121, | |
| "grad_norm": 1.6597908735275269, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 5.3542, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.004104507064488121, | |
| "eval_loss": 1.340001106262207, | |
| "eval_runtime": 73.6832, | |
| "eval_samples_per_second": 18.105, | |
| "eval_steps_per_second": 9.052, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.004420238377141053, | |
| "grad_norm": 2.3956403732299805, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 5.4276, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.004735969689793985, | |
| "grad_norm": 2.0606181621551514, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 5.2302, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.005051701002446917, | |
| "grad_norm": 2.8639981746673584, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 5.1071, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.00536743231509985, | |
| "grad_norm": 2.9613735675811768, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 4.8221, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.005683163627752783, | |
| "grad_norm": 2.5257747173309326, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 5.1752, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0059988949404057145, | |
| "grad_norm": 2.4979665279388428, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 5.448, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.006314626253058647, | |
| "grad_norm": 2.4641616344451904, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 4.5116, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.006630357565711579, | |
| "grad_norm": 2.583629608154297, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 4.3406, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.006946088878364512, | |
| "grad_norm": 2.122413158416748, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 4.1938, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0072618201910174445, | |
| "grad_norm": 1.9655640125274658, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 4.2416, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.007577551503670376, | |
| "grad_norm": 2.496213912963867, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 4.492, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.007893282816323309, | |
| "grad_norm": 2.098853349685669, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 4.0397, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.008209014128976242, | |
| "grad_norm": 2.3906338214874268, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 3.8275, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.008209014128976242, | |
| "eval_loss": 1.069798469543457, | |
| "eval_runtime": 73.654, | |
| "eval_samples_per_second": 18.112, | |
| "eval_steps_per_second": 9.056, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.008524745441629173, | |
| "grad_norm": 2.8270161151885986, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 4.3127, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.008840476754282105, | |
| "grad_norm": 2.343597173690796, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 4.7903, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.009156208066935038, | |
| "grad_norm": 2.1331675052642822, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 3.7894, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.00947193937958797, | |
| "grad_norm": 2.6220457553863525, | |
| "learning_rate": 0.0001, | |
| "loss": 4.2415, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.009787670692240904, | |
| "grad_norm": 2.525392532348633, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 4.3988, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.010103402004893835, | |
| "grad_norm": 2.459665298461914, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 4.202, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.010419133317546767, | |
| "grad_norm": 2.508035182952881, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 4.4926, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0107348646301997, | |
| "grad_norm": 2.238208293914795, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 4.3429, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.011050595942852633, | |
| "grad_norm": 1.8518725633621216, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 3.7004, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.011366327255505565, | |
| "grad_norm": 2.1632683277130127, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 3.6473, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.011682058568158496, | |
| "grad_norm": 2.4588687419891357, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 4.118, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.011997789880811429, | |
| "grad_norm": 1.9152288436889648, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 3.8266, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.012313521193464362, | |
| "grad_norm": 2.8986527919769287, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 4.5158, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.012313521193464362, | |
| "eval_loss": 0.9916089773178101, | |
| "eval_runtime": 73.6951, | |
| "eval_samples_per_second": 18.102, | |
| "eval_steps_per_second": 9.051, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.012629252506117294, | |
| "grad_norm": 2.411217212677002, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 4.3786, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.012944983818770227, | |
| "grad_norm": 1.5729023218154907, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 3.6625, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.013260715131423158, | |
| "grad_norm": 1.8918485641479492, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 3.5603, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.01357644644407609, | |
| "grad_norm": 2.446108818054199, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 3.903, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.013892177756729024, | |
| "grad_norm": 2.3776514530181885, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 3.9599, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.014207909069381956, | |
| "grad_norm": 2.1264560222625732, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 4.1211, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.014523640382034889, | |
| "grad_norm": 2.0649964809417725, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 4.0103, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.01483937169468782, | |
| "grad_norm": 2.101264238357544, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 3.6711, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.015155103007340753, | |
| "grad_norm": 2.364389419555664, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 3.9815, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.015470834319993685, | |
| "grad_norm": 2.1921024322509766, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 3.7505, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.015786565632646618, | |
| "grad_norm": 2.30208158493042, | |
| "learning_rate": 0.0, | |
| "loss": 3.8635, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.591027713441792e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |