| { |
| "best_global_step": 1400, |
| "best_metric": 82.55212784144531, |
| "best_model_checkpoint": null, |
| "epoch": 0.3973509933774834, |
| "eval_steps": 100, |
| "global_step": 1500, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.013245033112582781, |
| "grad_norm": 7.202547550201416, |
| "learning_rate": 9.998701986754968e-06, |
| "loss": 0.7094, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.026490066225165563, |
| "grad_norm": 9.662558555603027, |
| "learning_rate": 9.99737748344371e-06, |
| "loss": 0.68, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.026490066225165563, |
| "eval_HasAns_exact": 75.14792899408285, |
| "eval_HasAns_f1": 81.01957227969451, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 75.14792899408285, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 81.01957227969451, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 75.14792899408285, |
| "eval_f1": 81.01957227969451, |
| "eval_loss": 0.5824803113937378, |
| "eval_runtime": 120.1578, |
| "eval_samples_per_second": 125.177, |
| "eval_steps_per_second": 7.831, |
| "eval_total": 15041, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.039735099337748346, |
| "grad_norm": 7.527211666107178, |
| "learning_rate": 9.996052980132451e-06, |
| "loss": 0.586, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.052980132450331126, |
| "grad_norm": 5.419323921203613, |
| "learning_rate": 9.994728476821193e-06, |
| "loss": 0.6035, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.052980132450331126, |
| "eval_HasAns_exact": 75.56013562927997, |
| "eval_HasAns_f1": 81.46153699536843, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 75.56013562927997, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 81.46153699536843, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 75.56013562927997, |
| "eval_f1": 81.46153699536843, |
| "eval_loss": 0.5453085899353027, |
| "eval_runtime": 118.3719, |
| "eval_samples_per_second": 127.066, |
| "eval_steps_per_second": 7.95, |
| "eval_total": 15041, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.06622516556291391, |
| "grad_norm": 9.9099760055542, |
| "learning_rate": 9.993403973509935e-06, |
| "loss": 0.5616, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.07947019867549669, |
| "grad_norm": 3.5184719562530518, |
| "learning_rate": 9.992079470198677e-06, |
| "loss": 0.5286, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.07947019867549669, |
| "eval_HasAns_exact": 75.8526693703876, |
| "eval_HasAns_f1": 81.45745729278944, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 75.8526693703876, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 81.45745729278944, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 75.8526693703876, |
| "eval_f1": 81.45745729278944, |
| "eval_loss": 0.5200360417366028, |
| "eval_runtime": 118.7441, |
| "eval_samples_per_second": 126.667, |
| "eval_steps_per_second": 7.925, |
| "eval_total": 15041, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.09271523178807947, |
| "grad_norm": 4.889588832855225, |
| "learning_rate": 9.990754966887418e-06, |
| "loss": 0.4943, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.10596026490066225, |
| "grad_norm": 6.106881141662598, |
| "learning_rate": 9.98943046357616e-06, |
| "loss": 0.5957, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.10596026490066225, |
| "eval_HasAns_exact": 76.39784588790639, |
| "eval_HasAns_f1": 81.91199737367272, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 76.39784588790639, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 81.91199737367272, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 76.39784588790639, |
| "eval_f1": 81.91199737367272, |
| "eval_loss": 0.49782633781433105, |
| "eval_runtime": 118.2655, |
| "eval_samples_per_second": 127.18, |
| "eval_steps_per_second": 7.957, |
| "eval_total": 15041, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.11920529801324503, |
| "grad_norm": 6.963775634765625, |
| "learning_rate": 9.988105960264902e-06, |
| "loss": 0.5675, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.13245033112582782, |
| "grad_norm": 7.559121608734131, |
| "learning_rate": 9.986781456953643e-06, |
| "loss": 0.5173, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.13245033112582782, |
| "eval_HasAns_exact": 75.94574828801277, |
| "eval_HasAns_f1": 81.43615773338746, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 75.94574828801277, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 81.43615773338746, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 75.94574828801277, |
| "eval_f1": 81.43615773338746, |
| "eval_loss": 0.4918788969516754, |
| "eval_runtime": 118.4997, |
| "eval_samples_per_second": 126.929, |
| "eval_steps_per_second": 7.941, |
| "eval_total": 15041, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.1456953642384106, |
| "grad_norm": 5.2273268699646, |
| "learning_rate": 9.985456953642385e-06, |
| "loss": 0.5059, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.15894039735099338, |
| "grad_norm": 13.281010627746582, |
| "learning_rate": 9.984132450331127e-06, |
| "loss": 0.4847, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.15894039735099338, |
| "eval_HasAns_exact": 76.42443986437073, |
| "eval_HasAns_f1": 81.98556495448281, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 76.42443986437073, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 81.98556495448281, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 76.42443986437073, |
| "eval_f1": 81.98556495448281, |
| "eval_loss": 0.48446840047836304, |
| "eval_runtime": 118.3928, |
| "eval_samples_per_second": 127.043, |
| "eval_steps_per_second": 7.948, |
| "eval_total": 15041, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.17218543046357615, |
| "grad_norm": 18.346189498901367, |
| "learning_rate": 9.982807947019869e-06, |
| "loss": 0.4857, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.18543046357615894, |
| "grad_norm": 11.825461387634277, |
| "learning_rate": 9.98148344370861e-06, |
| "loss": 0.4764, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.18543046357615894, |
| "eval_HasAns_exact": 76.54411275846022, |
| "eval_HasAns_f1": 82.14175205695277, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 76.54411275846022, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 82.14175205695277, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 76.54411275846022, |
| "eval_f1": 82.14175205695277, |
| "eval_loss": 0.4717336595058441, |
| "eval_runtime": 118.5022, |
| "eval_samples_per_second": 126.926, |
| "eval_steps_per_second": 7.941, |
| "eval_total": 15041, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.1986754966887417, |
| "grad_norm": 7.6593475341796875, |
| "learning_rate": 9.980158940397352e-06, |
| "loss": 0.486, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.2119205298013245, |
| "grad_norm": 6.091643333435059, |
| "learning_rate": 9.978834437086094e-06, |
| "loss": 0.4707, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.2119205298013245, |
| "eval_HasAns_exact": 76.86324047603217, |
| "eval_HasAns_f1": 82.2239211673303, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 76.86324047603217, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 82.2239211673303, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 76.86324047603217, |
| "eval_f1": 82.2239211673303, |
| "eval_loss": 0.4601329565048218, |
| "eval_runtime": 119.1574, |
| "eval_samples_per_second": 126.228, |
| "eval_steps_per_second": 7.897, |
| "eval_total": 15041, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.2251655629139073, |
| "grad_norm": 4.863158702850342, |
| "learning_rate": 9.977509933774835e-06, |
| "loss": 0.5042, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.23841059602649006, |
| "grad_norm": 5.747822284698486, |
| "learning_rate": 9.976185430463577e-06, |
| "loss": 0.4725, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.23841059602649006, |
| "eval_HasAns_exact": 76.98291337012166, |
| "eval_HasAns_f1": 82.39127215176644, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 76.98291337012166, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 82.39127215176644, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 76.98291337012166, |
| "eval_f1": 82.39127215176644, |
| "eval_loss": 0.45768168568611145, |
| "eval_runtime": 118.7264, |
| "eval_samples_per_second": 126.686, |
| "eval_steps_per_second": 7.926, |
| "eval_total": 15041, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.25165562913907286, |
| "grad_norm": 14.890397071838379, |
| "learning_rate": 9.974860927152319e-06, |
| "loss": 0.4476, |
| "step": 950 |
| }, |
| { |
| "epoch": 0.26490066225165565, |
| "grad_norm": 8.837732315063477, |
| "learning_rate": 9.97353642384106e-06, |
| "loss": 0.4905, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.26490066225165565, |
| "eval_HasAns_exact": 77.10923475832723, |
| "eval_HasAns_f1": 82.37799175072134, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 77.10923475832723, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 82.37799175072134, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 77.10923475832723, |
| "eval_f1": 82.37799175072134, |
| "eval_loss": 0.44477367401123047, |
| "eval_runtime": 118.302, |
| "eval_samples_per_second": 127.141, |
| "eval_steps_per_second": 7.954, |
| "eval_total": 15041, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.2781456953642384, |
| "grad_norm": 5.750672817230225, |
| "learning_rate": 9.972211920529802e-06, |
| "loss": 0.5041, |
| "step": 1050 |
| }, |
| { |
| "epoch": 0.2913907284768212, |
| "grad_norm": 5.720890998840332, |
| "learning_rate": 9.970887417218544e-06, |
| "loss": 0.5156, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.2913907284768212, |
| "eval_HasAns_exact": 77.0826407818629, |
| "eval_HasAns_f1": 82.36916366838726, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 77.0826407818629, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 82.36916366838726, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 77.0826407818629, |
| "eval_f1": 82.36916366838726, |
| "eval_loss": 0.43873801827430725, |
| "eval_runtime": 118.2856, |
| "eval_samples_per_second": 127.158, |
| "eval_steps_per_second": 7.955, |
| "eval_total": 15041, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.304635761589404, |
| "grad_norm": 6.990389823913574, |
| "learning_rate": 9.969562913907286e-06, |
| "loss": 0.4507, |
| "step": 1150 |
| }, |
| { |
| "epoch": 0.31788079470198677, |
| "grad_norm": 11.490405082702637, |
| "learning_rate": 9.968238410596027e-06, |
| "loss": 0.4482, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.31788079470198677, |
| "eval_HasAns_exact": 77.01615584070208, |
| "eval_HasAns_f1": 82.29906445927755, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 77.01615584070208, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 82.29906445927755, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 77.01615584070208, |
| "eval_f1": 82.29906445927755, |
| "eval_loss": 0.42996492981910706, |
| "eval_runtime": 118.4966, |
| "eval_samples_per_second": 126.932, |
| "eval_steps_per_second": 7.941, |
| "eval_total": 15041, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.33112582781456956, |
| "grad_norm": 2.8688676357269287, |
| "learning_rate": 9.966913907284769e-06, |
| "loss": 0.4593, |
| "step": 1250 |
| }, |
| { |
| "epoch": 0.3443708609271523, |
| "grad_norm": 7.523802280426025, |
| "learning_rate": 9.96558940397351e-06, |
| "loss": 0.4548, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.3443708609271523, |
| "eval_HasAns_exact": 76.94967089954126, |
| "eval_HasAns_f1": 82.17470826997452, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 76.94967089954126, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 82.17470826997452, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 76.94967089954126, |
| "eval_f1": 82.17470826997452, |
| "eval_loss": 0.4349832832813263, |
| "eval_runtime": 118.6592, |
| "eval_samples_per_second": 126.758, |
| "eval_steps_per_second": 7.93, |
| "eval_total": 15041, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.3576158940397351, |
| "grad_norm": 10.102520942687988, |
| "learning_rate": 9.964264900662252e-06, |
| "loss": 0.4935, |
| "step": 1350 |
| }, |
| { |
| "epoch": 0.3708609271523179, |
| "grad_norm": 8.859588623046875, |
| "learning_rate": 9.962940397350994e-06, |
| "loss": 0.4344, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.3708609271523179, |
| "eval_HasAns_exact": 77.19566518183632, |
| "eval_HasAns_f1": 82.55212784144531, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 77.19566518183632, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 82.55212784144531, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 77.19566518183632, |
| "eval_f1": 82.55212784144531, |
| "eval_loss": 0.42366155982017517, |
| "eval_runtime": 119.1071, |
| "eval_samples_per_second": 126.281, |
| "eval_steps_per_second": 7.9, |
| "eval_total": 15041, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.3841059602649007, |
| "grad_norm": 5.980023384094238, |
| "learning_rate": 9.961615894039736e-06, |
| "loss": 0.4936, |
| "step": 1450 |
| }, |
| { |
| "epoch": 0.3973509933774834, |
| "grad_norm": 5.025954246520996, |
| "learning_rate": 9.960291390728478e-06, |
| "loss": 0.4431, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.3973509933774834, |
| "eval_HasAns_exact": 77.10923475832723, |
| "eval_HasAns_f1": 82.41666220398032, |
| "eval_HasAns_total": 15041, |
| "eval_best_exact": 77.10923475832723, |
| "eval_best_exact_thresh": 0.0, |
| "eval_best_f1": 82.41666220398032, |
| "eval_best_f1_thresh": 0.0, |
| "eval_exact": 77.10923475832723, |
| "eval_f1": 82.41666220398032, |
| "eval_loss": 0.4237441122531891, |
| "eval_runtime": 118.7094, |
| "eval_samples_per_second": 126.704, |
| "eval_steps_per_second": 7.927, |
| "eval_total": 15041, |
| "step": 1500 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 377500, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 100, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 10, |
| "early_stopping_threshold": 0.0 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 1 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.2542244323328e+16, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
