| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.23391812865497075, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004678362573099415, | |
| "grad_norm": 7.491190433502197, | |
| "learning_rate": 2e-05, | |
| "loss": 9.0167, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004678362573099415, | |
| "eval_loss": 2.5160412788391113, | |
| "eval_runtime": 17.8055, | |
| "eval_samples_per_second": 5.055, | |
| "eval_steps_per_second": 2.527, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00935672514619883, | |
| "grad_norm": 5.184515953063965, | |
| "learning_rate": 4e-05, | |
| "loss": 10.2909, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.014035087719298246, | |
| "grad_norm": 6.722748279571533, | |
| "learning_rate": 6e-05, | |
| "loss": 9.3702, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01871345029239766, | |
| "grad_norm": 8.923383712768555, | |
| "learning_rate": 8e-05, | |
| "loss": 9.7635, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.023391812865497075, | |
| "grad_norm": 7.835958957672119, | |
| "learning_rate": 0.0001, | |
| "loss": 7.9813, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.023391812865497075, | |
| "eval_loss": 2.45965576171875, | |
| "eval_runtime": 18.076, | |
| "eval_samples_per_second": 4.979, | |
| "eval_steps_per_second": 2.489, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.028070175438596492, | |
| "grad_norm": 5.639286994934082, | |
| "learning_rate": 0.00012, | |
| "loss": 8.7333, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.03274853801169591, | |
| "grad_norm": 9.286252975463867, | |
| "learning_rate": 0.00014, | |
| "loss": 7.7084, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.03742690058479532, | |
| "grad_norm": 11.736117362976074, | |
| "learning_rate": 0.00016, | |
| "loss": 7.0372, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.042105263157894736, | |
| "grad_norm": 13.311347007751465, | |
| "learning_rate": 0.00018, | |
| "loss": 7.8147, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04678362573099415, | |
| "grad_norm": 9.228219985961914, | |
| "learning_rate": 0.0002, | |
| "loss": 7.3818, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04678362573099415, | |
| "eval_loss": 1.8491570949554443, | |
| "eval_runtime": 18.1952, | |
| "eval_samples_per_second": 4.946, | |
| "eval_steps_per_second": 2.473, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05146198830409357, | |
| "grad_norm": 8.82361888885498, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 6.0434, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.056140350877192984, | |
| "grad_norm": 9.21557331085205, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 6.7836, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0608187134502924, | |
| "grad_norm": 6.748077869415283, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 5.2395, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.06549707602339182, | |
| "grad_norm": 8.914790153503418, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 4.4854, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.07017543859649122, | |
| "grad_norm": 9.942696571350098, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 4.5428, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.07017543859649122, | |
| "eval_loss": 0.8182364702224731, | |
| "eval_runtime": 18.2011, | |
| "eval_samples_per_second": 4.945, | |
| "eval_steps_per_second": 2.472, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.07485380116959064, | |
| "grad_norm": 12.87550163269043, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 3.2617, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.07953216374269007, | |
| "grad_norm": 13.529743194580078, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 2.5824, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.08421052631578947, | |
| "grad_norm": 15.365700721740723, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 2.4289, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.08888888888888889, | |
| "grad_norm": 9.652403831481934, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 2.119, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0935672514619883, | |
| "grad_norm": 9.246869087219238, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 1.8211, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0935672514619883, | |
| "eval_loss": 0.2940444350242615, | |
| "eval_runtime": 18.2125, | |
| "eval_samples_per_second": 4.942, | |
| "eval_steps_per_second": 2.471, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09824561403508772, | |
| "grad_norm": 8.619702339172363, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 1.3688, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.10292397660818714, | |
| "grad_norm": 5.638818740844727, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 1.1353, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.10760233918128655, | |
| "grad_norm": 10.630864143371582, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 1.3215, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.11228070175438597, | |
| "grad_norm": 6.898185729980469, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.825, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.11695906432748537, | |
| "grad_norm": 9.961976051330566, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 1.0316, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.11695906432748537, | |
| "eval_loss": 0.20844915509223938, | |
| "eval_runtime": 18.2036, | |
| "eval_samples_per_second": 4.944, | |
| "eval_steps_per_second": 2.472, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1216374269005848, | |
| "grad_norm": 4.300241470336914, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.6208, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.12631578947368421, | |
| "grad_norm": 6.638580799102783, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 1.0196, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.13099415204678364, | |
| "grad_norm": 4.728355407714844, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.6752, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.13567251461988303, | |
| "grad_norm": 11.99549388885498, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 1.2496, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.14035087719298245, | |
| "grad_norm": 5.10395622253418, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6347, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.14035087719298245, | |
| "eval_loss": 0.14852504432201385, | |
| "eval_runtime": 18.2195, | |
| "eval_samples_per_second": 4.94, | |
| "eval_steps_per_second": 2.47, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.14502923976608187, | |
| "grad_norm": 5.045573711395264, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.6205, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.1497076023391813, | |
| "grad_norm": 4.814074993133545, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.5803, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.1543859649122807, | |
| "grad_norm": 5.146168231964111, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.654, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.15906432748538013, | |
| "grad_norm": 4.554825782775879, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.6283, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.16374269005847952, | |
| "grad_norm": 3.142378091812134, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.5765, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.16374269005847952, | |
| "eval_loss": 0.1538289189338684, | |
| "eval_runtime": 18.2126, | |
| "eval_samples_per_second": 4.942, | |
| "eval_steps_per_second": 2.471, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.16842105263157894, | |
| "grad_norm": 4.004931449890137, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.6483, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.17309941520467836, | |
| "grad_norm": 5.645519256591797, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.6332, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.17777777777777778, | |
| "grad_norm": 3.922445297241211, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.5117, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.1824561403508772, | |
| "grad_norm": 6.259653568267822, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.6414, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1871345029239766, | |
| "grad_norm": 3.989673614501953, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.5892, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1871345029239766, | |
| "eval_loss": 0.142628476023674, | |
| "eval_runtime": 18.2463, | |
| "eval_samples_per_second": 4.933, | |
| "eval_steps_per_second": 2.466, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.19181286549707602, | |
| "grad_norm": 4.35756778717041, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.6447, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.19649122807017544, | |
| "grad_norm": 3.062830924987793, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.512, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.20116959064327486, | |
| "grad_norm": 3.9414889812469482, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.622, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.20584795321637428, | |
| "grad_norm": 3.3917744159698486, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.5016, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "grad_norm": 3.1187446117401123, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.5079, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "eval_loss": 0.13704170286655426, | |
| "eval_runtime": 18.2356, | |
| "eval_samples_per_second": 4.935, | |
| "eval_steps_per_second": 2.468, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2152046783625731, | |
| "grad_norm": 3.936415672302246, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.5366, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.2198830409356725, | |
| "grad_norm": 3.2776553630828857, | |
| "learning_rate": 2.7630079602323442e-06, | |
| "loss": 0.4972, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.22456140350877193, | |
| "grad_norm": 4.013199806213379, | |
| "learning_rate": 1.231165940486234e-06, | |
| "loss": 0.5533, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.22923976608187135, | |
| "grad_norm": 3.536522150039673, | |
| "learning_rate": 3.0826662668720364e-07, | |
| "loss": 0.51, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.23391812865497075, | |
| "grad_norm": 3.25169038772583, | |
| "learning_rate": 0.0, | |
| "loss": 0.4535, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.23391812865497075, | |
| "eval_loss": 0.13723856210708618, | |
| "eval_runtime": 18.2244, | |
| "eval_samples_per_second": 4.938, | |
| "eval_steps_per_second": 2.469, | |
| "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": 3.25746627182592e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |