| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.5789473684210527, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.031578947368421054, | |
| "grad_norm": 1.4201377630233765, | |
| "learning_rate": 2e-05, | |
| "loss": 1.2567, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.031578947368421054, | |
| "eval_loss": 1.2329180240631104, | |
| "eval_runtime": 9.7426, | |
| "eval_samples_per_second": 4.106, | |
| "eval_steps_per_second": 1.026, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06315789473684211, | |
| "grad_norm": 1.568509817123413, | |
| "learning_rate": 4e-05, | |
| "loss": 1.3621, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.09473684210526316, | |
| "grad_norm": 1.3656419515609741, | |
| "learning_rate": 6e-05, | |
| "loss": 1.3324, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.12631578947368421, | |
| "grad_norm": 1.3824914693832397, | |
| "learning_rate": 8e-05, | |
| "loss": 1.2106, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.15789473684210525, | |
| "grad_norm": 1.1639671325683594, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0618, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.15789473684210525, | |
| "eval_loss": 1.0954365730285645, | |
| "eval_runtime": 9.9704, | |
| "eval_samples_per_second": 4.012, | |
| "eval_steps_per_second": 1.003, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.18947368421052632, | |
| "grad_norm": 1.0370177030563354, | |
| "learning_rate": 0.00012, | |
| "loss": 1.0542, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.22105263157894736, | |
| "grad_norm": 0.7691447138786316, | |
| "learning_rate": 0.00014, | |
| "loss": 1.0324, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.25263157894736843, | |
| "grad_norm": 0.8053244948387146, | |
| "learning_rate": 0.00016, | |
| "loss": 1.0138, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.28421052631578947, | |
| "grad_norm": 1.1447694301605225, | |
| "learning_rate": 0.00018, | |
| "loss": 1.0227, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "grad_norm": 0.7161327004432678, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8714, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "eval_loss": 0.8887165188789368, | |
| "eval_runtime": 9.9898, | |
| "eval_samples_per_second": 4.004, | |
| "eval_steps_per_second": 1.001, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3473684210526316, | |
| "grad_norm": 0.8024160861968994, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 0.87, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.37894736842105264, | |
| "grad_norm": 0.6736723184585571, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 0.9436, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.4105263157894737, | |
| "grad_norm": 0.4639568626880646, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.9028, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.4421052631578947, | |
| "grad_norm": 0.4748899042606354, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.8735, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "grad_norm": 0.5926955938339233, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.9889, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.47368421052631576, | |
| "eval_loss": 0.810930609703064, | |
| "eval_runtime": 9.9943, | |
| "eval_samples_per_second": 4.002, | |
| "eval_steps_per_second": 1.001, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5052631578947369, | |
| "grad_norm": 0.4821624755859375, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.8502, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.5368421052631579, | |
| "grad_norm": 0.43275609612464905, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.7635, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.5684210526315789, | |
| "grad_norm": 0.5297432541847229, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.7952, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.5860052704811096, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 1.0571, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "grad_norm": 0.47979608178138733, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.7376, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "eval_loss": 0.779615044593811, | |
| "eval_runtime": 9.9911, | |
| "eval_samples_per_second": 4.004, | |
| "eval_steps_per_second": 1.001, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6631578947368421, | |
| "grad_norm": 0.36290550231933594, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.7565, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.6947368421052632, | |
| "grad_norm": 0.43947532773017883, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.7572, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.7263157894736842, | |
| "grad_norm": 0.7117337584495544, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.8235, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.7578947368421053, | |
| "grad_norm": 0.41661393642425537, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.7186, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.7894736842105263, | |
| "grad_norm": 0.43488454818725586, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.8079, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.7894736842105263, | |
| "eval_loss": 0.7661096453666687, | |
| "eval_runtime": 9.9986, | |
| "eval_samples_per_second": 4.001, | |
| "eval_steps_per_second": 1.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.8210526315789474, | |
| "grad_norm": 0.5269589424133301, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.9215, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.8526315789473684, | |
| "grad_norm": 0.43179982900619507, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.7455, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.8842105263157894, | |
| "grad_norm": 0.5215819478034973, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.861, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.9157894736842105, | |
| "grad_norm": 0.3976958692073822, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.7871, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "grad_norm": 0.41575056314468384, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7532, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "eval_loss": 0.7560582160949707, | |
| "eval_runtime": 9.9942, | |
| "eval_samples_per_second": 4.002, | |
| "eval_steps_per_second": 1.001, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9789473684210527, | |
| "grad_norm": 0.49081918597221375, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.6406, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.0105263157894737, | |
| "grad_norm": 0.5961962938308716, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 1.0858, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.0421052631578946, | |
| "grad_norm": 0.3866117596626282, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.7171, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.0736842105263158, | |
| "grad_norm": 0.4523923695087433, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.6907, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.1052631578947367, | |
| "grad_norm": 0.3665292263031006, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.7877, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.1052631578947367, | |
| "eval_loss": 0.7507282495498657, | |
| "eval_runtime": 9.999, | |
| "eval_samples_per_second": 4.0, | |
| "eval_steps_per_second": 1.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.1368421052631579, | |
| "grad_norm": 0.3150940239429474, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.61, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.168421052631579, | |
| "grad_norm": 0.39909401535987854, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.7461, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.3504452705383301, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.7198, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.231578947368421, | |
| "grad_norm": 0.3640637993812561, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.5523, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.263157894736842, | |
| "grad_norm": 0.3753401041030884, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.7139, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.263157894736842, | |
| "eval_loss": 0.752138614654541, | |
| "eval_runtime": 9.9936, | |
| "eval_samples_per_second": 4.003, | |
| "eval_steps_per_second": 1.001, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.2947368421052632, | |
| "grad_norm": 0.3601594865322113, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.6394, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.3263157894736843, | |
| "grad_norm": 0.40832358598709106, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.8728, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.3578947368421053, | |
| "grad_norm": 0.44947680830955505, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.9453, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.3894736842105262, | |
| "grad_norm": 0.3563525080680847, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.5642, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.4210526315789473, | |
| "grad_norm": 0.4350314438343048, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.7842, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.4210526315789473, | |
| "eval_loss": 0.7522160410881042, | |
| "eval_runtime": 9.9932, | |
| "eval_samples_per_second": 4.003, | |
| "eval_steps_per_second": 1.001, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.4526315789473685, | |
| "grad_norm": 0.3606291711330414, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.5951, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.4842105263157894, | |
| "grad_norm": 0.35893508791923523, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.7562, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.5157894736842106, | |
| "grad_norm": 0.4855940341949463, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.9173, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.5473684210526315, | |
| "grad_norm": 0.48101991415023804, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.8475, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.5789473684210527, | |
| "grad_norm": 0.3735427260398865, | |
| "learning_rate": 0.0, | |
| "loss": 0.8305, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.5789473684210527, | |
| "eval_loss": 0.7518041729927063, | |
| "eval_runtime": 9.9913, | |
| "eval_samples_per_second": 4.003, | |
| "eval_steps_per_second": 1.001, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.112830925340672e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |