| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0012921482610914776, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.5842965221829554e-05, | |
| "grad_norm": 3.3958845138549805, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4244, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.5842965221829554e-05, | |
| "eval_loss": 2.1125328540802, | |
| "eval_runtime": 1391.8533, | |
| "eval_samples_per_second": 11.706, | |
| "eval_steps_per_second": 5.853, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 5.168593044365911e-05, | |
| "grad_norm": 2.459746837615967, | |
| "learning_rate": 4e-05, | |
| "loss": 1.5737, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 7.752889566548866e-05, | |
| "grad_norm": 2.5988481044769287, | |
| "learning_rate": 6e-05, | |
| "loss": 1.2215, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00010337186088731822, | |
| "grad_norm": 2.9697864055633545, | |
| "learning_rate": 8e-05, | |
| "loss": 1.3948, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00012921482610914775, | |
| "grad_norm": 3.4510738849639893, | |
| "learning_rate": 0.0001, | |
| "loss": 3.3687, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00012921482610914775, | |
| "eval_loss": 1.8981174230575562, | |
| "eval_runtime": 1395.5569, | |
| "eval_samples_per_second": 11.675, | |
| "eval_steps_per_second": 5.838, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0001550577913309773, | |
| "grad_norm": 2.285410165786743, | |
| "learning_rate": 0.00012, | |
| "loss": 1.0475, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00018090075655280687, | |
| "grad_norm": 3.462702751159668, | |
| "learning_rate": 0.00014, | |
| "loss": 1.0077, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00020674372177463643, | |
| "grad_norm": 2.645705223083496, | |
| "learning_rate": 0.00016, | |
| "loss": 0.8847, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00023258668699646597, | |
| "grad_norm": 2.0876657962799072, | |
| "learning_rate": 0.00018, | |
| "loss": 0.7847, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0002584296522182955, | |
| "grad_norm": 2.068669319152832, | |
| "learning_rate": 0.0002, | |
| "loss": 0.6889, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0002584296522182955, | |
| "eval_loss": 1.0108299255371094, | |
| "eval_runtime": 1395.6064, | |
| "eval_samples_per_second": 11.674, | |
| "eval_steps_per_second": 5.838, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0002842726174401251, | |
| "grad_norm": 1.778637170791626, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 0.6103, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0003101155826619546, | |
| "grad_norm": 1.7560285329818726, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.4017, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0003359585478837842, | |
| "grad_norm": 1.798984169960022, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.4021, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00036180151310561374, | |
| "grad_norm": 2.582660436630249, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.4864, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0003876444783274433, | |
| "grad_norm": 2.107647180557251, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.3102, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0003876444783274433, | |
| "eval_loss": 0.4638846814632416, | |
| "eval_runtime": 1395.4516, | |
| "eval_samples_per_second": 11.676, | |
| "eval_steps_per_second": 5.838, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00041348744354927287, | |
| "grad_norm": 2.237151622772217, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.3547, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0004393304087711024, | |
| "grad_norm": 2.265132427215576, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.3212, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.00046517337399293193, | |
| "grad_norm": 3.5674264430999756, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.3152, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0004910163392147615, | |
| "grad_norm": 1.4233182668685913, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.2144, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.000516859304436591, | |
| "grad_norm": 1.286632776260376, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.1448, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.000516859304436591, | |
| "eval_loss": 0.27176007628440857, | |
| "eval_runtime": 1395.268, | |
| "eval_samples_per_second": 11.677, | |
| "eval_steps_per_second": 5.839, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0005427022696584206, | |
| "grad_norm": 1.321890950202942, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.1948, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0005685452348802502, | |
| "grad_norm": 2.3737189769744873, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.2611, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0005943882001020797, | |
| "grad_norm": 1.2037267684936523, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.1581, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0006202311653239092, | |
| "grad_norm": 1.1055679321289062, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.1362, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0006460741305457388, | |
| "grad_norm": 0.9288361668586731, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.1606, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0006460741305457388, | |
| "eval_loss": 0.19134792685508728, | |
| "eval_runtime": 1395.0425, | |
| "eval_samples_per_second": 11.679, | |
| "eval_steps_per_second": 5.84, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0006719170957675684, | |
| "grad_norm": 0.7735410928726196, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.1043, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.000697760060989398, | |
| "grad_norm": 1.195586085319519, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.1089, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0007236030262112275, | |
| "grad_norm": 1.4720776081085205, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.173, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.000749445991433057, | |
| "grad_norm": 1.69827401638031, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.1277, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0007752889566548866, | |
| "grad_norm": 0.697060227394104, | |
| "learning_rate": 0.0001, | |
| "loss": 0.104, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0007752889566548866, | |
| "eval_loss": 0.16827471554279327, | |
| "eval_runtime": 1395.0717, | |
| "eval_samples_per_second": 11.679, | |
| "eval_steps_per_second": 5.84, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0008011319218767161, | |
| "grad_norm": 1.0948371887207031, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.1049, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0008269748870985457, | |
| "grad_norm": 2.184497594833374, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.1007, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0008528178523203753, | |
| "grad_norm": 1.1975798606872559, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.1052, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0008786608175422048, | |
| "grad_norm": 1.8608812093734741, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.1533, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0009045037827640343, | |
| "grad_norm": 0.8412917256355286, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.0917, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0009045037827640343, | |
| "eval_loss": 0.15119507908821106, | |
| "eval_runtime": 1394.8383, | |
| "eval_samples_per_second": 11.681, | |
| "eval_steps_per_second": 5.841, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0009303467479858639, | |
| "grad_norm": 0.6619991064071655, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.1093, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0009561897132076934, | |
| "grad_norm": 0.6209729313850403, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.06, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.000982032678429523, | |
| "grad_norm": 0.650658905506134, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.0841, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0010078756436513526, | |
| "grad_norm": 0.9135183095932007, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.1116, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.001033718608873182, | |
| "grad_norm": 0.8710368275642395, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.0838, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.001033718608873182, | |
| "eval_loss": 0.142450749874115, | |
| "eval_runtime": 1394.9906, | |
| "eval_samples_per_second": 11.68, | |
| "eval_steps_per_second": 5.84, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0010595615740950116, | |
| "grad_norm": 0.9050641655921936, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.1035, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0010854045393168413, | |
| "grad_norm": 0.877323567867279, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.1082, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0011112475045386707, | |
| "grad_norm": 1.5448343753814697, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.1315, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0011370904697605004, | |
| "grad_norm": 0.8033886551856995, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.0732, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0011629334349823298, | |
| "grad_norm": 1.3498198986053467, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.1228, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0011629334349823298, | |
| "eval_loss": 0.1376151740550995, | |
| "eval_runtime": 1395.1018, | |
| "eval_samples_per_second": 11.679, | |
| "eval_steps_per_second": 5.84, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0011887764002041594, | |
| "grad_norm": 1.1750056743621826, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.1394, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.001214619365425989, | |
| "grad_norm": 2.630324363708496, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.0545, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0012404623306478185, | |
| "grad_norm": 0.5797359347343445, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.0958, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0012663052958696481, | |
| "grad_norm": 1.0320329666137695, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.0997, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0012921482610914776, | |
| "grad_norm": 0.5513184070587158, | |
| "learning_rate": 0.0, | |
| "loss": 0.0691, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0012921482610914776, | |
| "eval_loss": 0.1370484083890915, | |
| "eval_runtime": 1398.1486, | |
| "eval_samples_per_second": 11.653, | |
| "eval_steps_per_second": 5.827, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 70, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.84956424814592e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |