| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.8942544153811759, | |
| "eval_steps": 50, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.017885088307623517, | |
| "grad_norm": 2.1336276531219482, | |
| "learning_rate": 2e-05, | |
| "loss": 2.0684, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.035770176615247035, | |
| "grad_norm": 1.5654758214950562, | |
| "learning_rate": 4e-05, | |
| "loss": 1.7055, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05365526492287056, | |
| "grad_norm": 1.2047432661056519, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9471, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07154035323049407, | |
| "grad_norm": 0.1950729787349701, | |
| "learning_rate": 8e-05, | |
| "loss": 0.1249, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0894254415381176, | |
| "grad_norm": 0.09746120870113373, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0457, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0894254415381176, | |
| "eval_loss": 0.04336774721741676, | |
| "eval_runtime": 342.8992, | |
| "eval_samples_per_second": 5.798, | |
| "eval_steps_per_second": 2.899, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.10731052984574112, | |
| "grad_norm": 0.10309002548456192, | |
| "learning_rate": 9.777777777777778e-05, | |
| "loss": 0.0417, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12519561815336464, | |
| "grad_norm": 0.10960160195827484, | |
| "learning_rate": 9.555555555555557e-05, | |
| "loss": 0.038, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.14308070646098814, | |
| "grad_norm": 0.1561357080936432, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 0.032, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.16096579476861167, | |
| "grad_norm": 0.0917748436331749, | |
| "learning_rate": 9.111111111111112e-05, | |
| "loss": 0.0213, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1788508830762352, | |
| "grad_norm": 0.06082587316632271, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.0156, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1788508830762352, | |
| "eval_loss": 0.010981782339513302, | |
| "eval_runtime": 343.0451, | |
| "eval_samples_per_second": 5.795, | |
| "eval_steps_per_second": 2.898, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1967359713838587, | |
| "grad_norm": 0.11666049808263779, | |
| "learning_rate": 8.666666666666667e-05, | |
| "loss": 0.0085, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.21462105969148224, | |
| "grad_norm": 0.05114075168967247, | |
| "learning_rate": 8.444444444444444e-05, | |
| "loss": 0.0038, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.23250614799910574, | |
| "grad_norm": 0.03657279163599014, | |
| "learning_rate": 8.222222222222222e-05, | |
| "loss": 0.0031, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.25039123630672927, | |
| "grad_norm": 0.013491852208971977, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0029, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.2682763246143528, | |
| "grad_norm": 0.006547242868691683, | |
| "learning_rate": 7.777777777777778e-05, | |
| "loss": 0.0029, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2682763246143528, | |
| "eval_loss": 0.002737525850534439, | |
| "eval_runtime": 342.9108, | |
| "eval_samples_per_second": 5.797, | |
| "eval_steps_per_second": 2.899, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2861614129219763, | |
| "grad_norm": 0.005285318940877914, | |
| "learning_rate": 7.555555555555556e-05, | |
| "loss": 0.0028, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.3040465012295998, | |
| "grad_norm": 0.006457226816564798, | |
| "learning_rate": 7.333333333333333e-05, | |
| "loss": 0.0028, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.32193158953722334, | |
| "grad_norm": 0.0023280009627342224, | |
| "learning_rate": 7.111111111111112e-05, | |
| "loss": 0.0027, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.33981667784484687, | |
| "grad_norm": 0.0014506684383377433, | |
| "learning_rate": 6.88888888888889e-05, | |
| "loss": 0.0028, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.3577017661524704, | |
| "grad_norm": 0.002698718337342143, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.0028, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3577017661524704, | |
| "eval_loss": 0.0027305446565151215, | |
| "eval_runtime": 342.6882, | |
| "eval_samples_per_second": 5.801, | |
| "eval_steps_per_second": 2.901, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3755868544600939, | |
| "grad_norm": 0.003260605735704303, | |
| "learning_rate": 6.444444444444446e-05, | |
| "loss": 0.0028, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.3934719427677174, | |
| "grad_norm": 0.002418739954009652, | |
| "learning_rate": 6.222222222222222e-05, | |
| "loss": 0.0028, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.41135703107534094, | |
| "grad_norm": 0.001470407354645431, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0028, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.42924211938296447, | |
| "grad_norm": 0.004997240845113993, | |
| "learning_rate": 5.7777777777777776e-05, | |
| "loss": 0.0027, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.44712720769058795, | |
| "grad_norm": 0.0029559540562331676, | |
| "learning_rate": 5.555555555555556e-05, | |
| "loss": 0.0028, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.44712720769058795, | |
| "eval_loss": 0.0027349325828254223, | |
| "eval_runtime": 342.8604, | |
| "eval_samples_per_second": 5.798, | |
| "eval_steps_per_second": 2.899, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4650122959982115, | |
| "grad_norm": 0.004728947766125202, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.0028, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.482897384305835, | |
| "grad_norm": 0.003956568893045187, | |
| "learning_rate": 5.111111111111111e-05, | |
| "loss": 0.0027, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.5007824726134585, | |
| "grad_norm": 0.0033387644216418266, | |
| "learning_rate": 4.888888888888889e-05, | |
| "loss": 0.0028, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.518667560921082, | |
| "grad_norm": 0.0044438643380999565, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.0028, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.5365526492287056, | |
| "grad_norm": 0.002528380835428834, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.0028, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5365526492287056, | |
| "eval_loss": 0.0027327719144523144, | |
| "eval_runtime": 342.796, | |
| "eval_samples_per_second": 5.799, | |
| "eval_steps_per_second": 2.9, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5544377375363291, | |
| "grad_norm": 0.003869276260957122, | |
| "learning_rate": 4.222222222222222e-05, | |
| "loss": 0.0028, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.5723228258439526, | |
| "grad_norm": 0.0063848476856946945, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0027, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.5902079141515761, | |
| "grad_norm": 0.00741184875369072, | |
| "learning_rate": 3.777777777777778e-05, | |
| "loss": 0.0027, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.6080930024591996, | |
| "grad_norm": 0.007532441522926092, | |
| "learning_rate": 3.555555555555556e-05, | |
| "loss": 0.0028, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.6259780907668232, | |
| "grad_norm": 0.0035441338550299406, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0028, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6259780907668232, | |
| "eval_loss": 0.00272406660951674, | |
| "eval_runtime": 342.7543, | |
| "eval_samples_per_second": 5.8, | |
| "eval_steps_per_second": 2.9, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6438631790744467, | |
| "grad_norm": 0.002798376139253378, | |
| "learning_rate": 3.111111111111111e-05, | |
| "loss": 0.0028, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.6617482673820702, | |
| "grad_norm": 0.0012276788474991918, | |
| "learning_rate": 2.8888888888888888e-05, | |
| "loss": 0.0028, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.6796333556896937, | |
| "grad_norm": 0.0022674233186990023, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.0027, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.6975184439973172, | |
| "grad_norm": 0.004862118978053331, | |
| "learning_rate": 2.4444444444444445e-05, | |
| "loss": 0.0028, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.7154035323049408, | |
| "grad_norm": 0.0016343879979103804, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0027, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7154035323049408, | |
| "eval_loss": 0.0027308915741741657, | |
| "eval_runtime": 343.0129, | |
| "eval_samples_per_second": 5.796, | |
| "eval_steps_per_second": 2.898, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7332886206125643, | |
| "grad_norm": 0.0016705788439139724, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0027, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.7511737089201878, | |
| "grad_norm": 0.0037323199212551117, | |
| "learning_rate": 1.777777777777778e-05, | |
| "loss": 0.0028, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.7690587972278113, | |
| "grad_norm": 0.002069295383989811, | |
| "learning_rate": 1.5555555555555555e-05, | |
| "loss": 0.0028, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.7869438855354348, | |
| "grad_norm": 0.00587103795260191, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.0027, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.8048289738430584, | |
| "grad_norm": 0.004777251742780209, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0027, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.8048289738430584, | |
| "eval_loss": 0.0027228216640651226, | |
| "eval_runtime": 342.6715, | |
| "eval_samples_per_second": 5.801, | |
| "eval_steps_per_second": 2.901, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.8227140621506819, | |
| "grad_norm": 0.004516826942563057, | |
| "learning_rate": 8.88888888888889e-06, | |
| "loss": 0.0028, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.8405991504583054, | |
| "grad_norm": 0.0023202409502118826, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.0027, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.8584842387659289, | |
| "grad_norm": 0.003505381755530834, | |
| "learning_rate": 4.444444444444445e-06, | |
| "loss": 0.0028, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.8763693270735524, | |
| "grad_norm": 0.004354626871645451, | |
| "learning_rate": 2.2222222222222225e-06, | |
| "loss": 0.0027, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.8942544153811759, | |
| "grad_norm": 0.0022386154159903526, | |
| "learning_rate": 0.0, | |
| "loss": 0.0027, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8942544153811759, | |
| "eval_loss": 0.0027302990201860666, | |
| "eval_runtime": 342.9424, | |
| "eval_samples_per_second": 5.797, | |
| "eval_steps_per_second": 2.898, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.710113009369088e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |