| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 8.02923264311815, |
| "eval_steps": 100, |
| "global_step": 3300, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.243605359317905, |
| "grad_norm": 33.11063766479492, |
| "learning_rate": 4.8844282238442824e-05, |
| "loss": 10.7343, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.243605359317905, |
| "eval_loss": 4.385446548461914, |
| "eval_runtime": 206.0799, |
| "eval_samples_per_second": 7.992, |
| "eval_steps_per_second": 1.0, |
| "eval_wer": 1.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.48721071863581, |
| "grad_norm": 2.508547067642212, |
| "learning_rate": 4.762773722627738e-05, |
| "loss": 3.1325, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.48721071863581, |
| "eval_loss": 2.1581695079803467, |
| "eval_runtime": 206.4525, |
| "eval_samples_per_second": 7.978, |
| "eval_steps_per_second": 0.998, |
| "eval_wer": 0.9690996270644646, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.730816077953715, |
| "grad_norm": 1.2504737377166748, |
| "learning_rate": 4.641119221411192e-05, |
| "loss": 1.6166, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.730816077953715, |
| "eval_loss": 1.1347323656082153, |
| "eval_runtime": 206.3117, |
| "eval_samples_per_second": 7.983, |
| "eval_steps_per_second": 0.998, |
| "eval_wer": 0.7188184455099745, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.97442143727162, |
| "grad_norm": 1.3984403610229492, |
| "learning_rate": 4.520681265206813e-05, |
| "loss": 1.1808, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.97442143727162, |
| "eval_loss": 0.9792075157165527, |
| "eval_runtime": 206.2345, |
| "eval_samples_per_second": 7.986, |
| "eval_steps_per_second": 0.999, |
| "eval_wer": 0.6746581424258569, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.2168087697929355, |
| "grad_norm": 10.769020080566406, |
| "learning_rate": 4.3990267639902676e-05, |
| "loss": 1.0459, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.2168087697929355, |
| "eval_loss": 0.9049620032310486, |
| "eval_runtime": 208.2322, |
| "eval_samples_per_second": 7.909, |
| "eval_steps_per_second": 0.989, |
| "eval_wer": 0.6494405966968567, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.4604141291108403, |
| "grad_norm": 0.8941505551338196, |
| "learning_rate": 4.277372262773723e-05, |
| "loss": 1.0317, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.4604141291108403, |
| "eval_loss": 0.8543379306793213, |
| "eval_runtime": 205.6459, |
| "eval_samples_per_second": 8.009, |
| "eval_steps_per_second": 1.002, |
| "eval_wer": 0.6338128218788848, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.7040194884287454, |
| "grad_norm": 0.9768113493919373, |
| "learning_rate": 4.1569343065693435e-05, |
| "loss": 0.9836, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.7040194884287454, |
| "eval_loss": 0.8191400766372681, |
| "eval_runtime": 207.1525, |
| "eval_samples_per_second": 7.951, |
| "eval_steps_per_second": 0.994, |
| "eval_wer": 0.6251701888356124, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.9476248477466505, |
| "grad_norm": 0.9090480208396912, |
| "learning_rate": 4.035279805352798e-05, |
| "loss": 0.9567, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.9476248477466505, |
| "eval_loss": 0.7955169081687927, |
| "eval_runtime": 206.8205, |
| "eval_samples_per_second": 7.963, |
| "eval_steps_per_second": 0.996, |
| "eval_wer": 0.6124430237376428, |
| "step": 800 |
| }, |
| { |
| "epoch": 2.1900121802679657, |
| "grad_norm": 0.6563196778297424, |
| "learning_rate": 3.913625304136253e-05, |
| "loss": 0.9354, |
| "step": 900 |
| }, |
| { |
| "epoch": 2.1900121802679657, |
| "eval_loss": 0.7705180644989014, |
| "eval_runtime": 206.4429, |
| "eval_samples_per_second": 7.978, |
| "eval_steps_per_second": 0.998, |
| "eval_wer": 0.6046291363286569, |
| "step": 900 |
| }, |
| { |
| "epoch": 2.433617539585871, |
| "grad_norm": 0.8123560547828674, |
| "learning_rate": 3.791970802919708e-05, |
| "loss": 0.9037, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.433617539585871, |
| "eval_loss": 0.75263512134552, |
| "eval_runtime": 206.6571, |
| "eval_samples_per_second": 7.97, |
| "eval_steps_per_second": 0.997, |
| "eval_wer": 0.5982359557213047, |
| "step": 1000 |
| }, |
| { |
| "epoch": 2.677222898903776, |
| "grad_norm": 0.8349791169166565, |
| "learning_rate": 3.6703163017031634e-05, |
| "loss": 0.901, |
| "step": 1100 |
| }, |
| { |
| "epoch": 2.677222898903776, |
| "eval_loss": 0.7370020747184753, |
| "eval_runtime": 207.9183, |
| "eval_samples_per_second": 7.921, |
| "eval_steps_per_second": 0.991, |
| "eval_wer": 0.5959865032853845, |
| "step": 1100 |
| }, |
| { |
| "epoch": 2.9208282582216807, |
| "grad_norm": 2.0188732147216797, |
| "learning_rate": 3.549878345498784e-05, |
| "loss": 0.8888, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.9208282582216807, |
| "eval_loss": 0.7251046895980835, |
| "eval_runtime": 208.4027, |
| "eval_samples_per_second": 7.903, |
| "eval_steps_per_second": 0.988, |
| "eval_wer": 0.58781743917599, |
| "step": 1200 |
| }, |
| { |
| "epoch": 3.1632155907429964, |
| "grad_norm": 0.4746805429458618, |
| "learning_rate": 3.4282238442822386e-05, |
| "loss": 0.8686, |
| "step": 1300 |
| }, |
| { |
| "epoch": 3.1632155907429964, |
| "eval_loss": 0.7125080823898315, |
| "eval_runtime": 206.121, |
| "eval_samples_per_second": 7.99, |
| "eval_steps_per_second": 0.999, |
| "eval_wer": 0.5834369265376191, |
| "step": 1300 |
| }, |
| { |
| "epoch": 3.4068209500609012, |
| "grad_norm": 1.0152044296264648, |
| "learning_rate": 3.306569343065693e-05, |
| "loss": 0.8681, |
| "step": 1400 |
| }, |
| { |
| "epoch": 3.4068209500609012, |
| "eval_loss": 0.7029690742492676, |
| "eval_runtime": 206.1729, |
| "eval_samples_per_second": 7.988, |
| "eval_steps_per_second": 0.999, |
| "eval_wer": 0.577043745930267, |
| "step": 1400 |
| }, |
| { |
| "epoch": 3.6504263093788065, |
| "grad_norm": 0.5681760907173157, |
| "learning_rate": 3.1849148418491486e-05, |
| "loss": 0.8428, |
| "step": 1500 |
| }, |
| { |
| "epoch": 3.6504263093788065, |
| "eval_loss": 0.6939159035682678, |
| "eval_runtime": 206.9126, |
| "eval_samples_per_second": 7.96, |
| "eval_steps_per_second": 0.996, |
| "eval_wer": 0.572900017758835, |
| "step": 1500 |
| }, |
| { |
| "epoch": 3.8940316686967114, |
| "grad_norm": 0.535625159740448, |
| "learning_rate": 3.064476885644769e-05, |
| "loss": 0.8372, |
| "step": 1600 |
| }, |
| { |
| "epoch": 3.8940316686967114, |
| "eval_loss": 0.6848719716072083, |
| "eval_runtime": 206.7992, |
| "eval_samples_per_second": 7.964, |
| "eval_steps_per_second": 0.996, |
| "eval_wer": 0.5706505653229148, |
| "step": 1600 |
| }, |
| { |
| "epoch": 4.136419001218027, |
| "grad_norm": 1.2212234735488892, |
| "learning_rate": 2.942822384428224e-05, |
| "loss": 0.8388, |
| "step": 1700 |
| }, |
| { |
| "epoch": 4.136419001218027, |
| "eval_loss": 0.6779345870018005, |
| "eval_runtime": 206.7198, |
| "eval_samples_per_second": 7.967, |
| "eval_steps_per_second": 0.997, |
| "eval_wer": 0.5667436216184218, |
| "step": 1700 |
| }, |
| { |
| "epoch": 4.3800243605359315, |
| "grad_norm": 0.515786349773407, |
| "learning_rate": 2.8211678832116788e-05, |
| "loss": 0.8222, |
| "step": 1800 |
| }, |
| { |
| "epoch": 4.3800243605359315, |
| "eval_loss": 0.6726759672164917, |
| "eval_runtime": 206.3937, |
| "eval_samples_per_second": 7.98, |
| "eval_steps_per_second": 0.998, |
| "eval_wer": 0.5620671283963772, |
| "step": 1800 |
| }, |
| { |
| "epoch": 4.623629719853836, |
| "grad_norm": 2.3206946849823, |
| "learning_rate": 2.699513381995134e-05, |
| "loss": 0.8289, |
| "step": 1900 |
| }, |
| { |
| "epoch": 4.623629719853836, |
| "eval_loss": 0.6664962768554688, |
| "eval_runtime": 206.3925, |
| "eval_samples_per_second": 7.98, |
| "eval_steps_per_second": 0.998, |
| "eval_wer": 0.5570354584739241, |
| "step": 1900 |
| }, |
| { |
| "epoch": 4.867235079171742, |
| "grad_norm": 2.1069090366363525, |
| "learning_rate": 2.5790754257907544e-05, |
| "loss": 0.8189, |
| "step": 2000 |
| }, |
| { |
| "epoch": 4.867235079171742, |
| "eval_loss": 0.6623280048370361, |
| "eval_runtime": 208.1032, |
| "eval_samples_per_second": 7.914, |
| "eval_steps_per_second": 0.99, |
| "eval_wer": 0.5564434973065767, |
| "step": 2000 |
| }, |
| { |
| "epoch": 5.109622411693057, |
| "grad_norm": 1.1477903127670288, |
| "learning_rate": 2.4574209245742094e-05, |
| "loss": 0.8073, |
| "step": 2100 |
| }, |
| { |
| "epoch": 5.109622411693057, |
| "eval_loss": 0.6582257151603699, |
| "eval_runtime": 206.4132, |
| "eval_samples_per_second": 7.979, |
| "eval_steps_per_second": 0.998, |
| "eval_wer": 0.5535428875865743, |
| "step": 2100 |
| }, |
| { |
| "epoch": 5.353227771010962, |
| "grad_norm": 1.1735588312149048, |
| "learning_rate": 2.3357664233576643e-05, |
| "loss": 0.8, |
| "step": 2200 |
| }, |
| { |
| "epoch": 5.353227771010962, |
| "eval_loss": 0.6531500816345215, |
| "eval_runtime": 206.8033, |
| "eval_samples_per_second": 7.964, |
| "eval_steps_per_second": 0.996, |
| "eval_wer": 0.5504646895163677, |
| "step": 2200 |
| }, |
| { |
| "epoch": 5.596833130328867, |
| "grad_norm": 0.8227632641792297, |
| "learning_rate": 2.2141119221411193e-05, |
| "loss": 0.8051, |
| "step": 2300 |
| }, |
| { |
| "epoch": 5.596833130328867, |
| "eval_loss": 0.648690402507782, |
| "eval_runtime": 208.3131, |
| "eval_samples_per_second": 7.906, |
| "eval_steps_per_second": 0.989, |
| "eval_wer": 0.5461433729947316, |
| "step": 2300 |
| }, |
| { |
| "epoch": 5.840438489646772, |
| "grad_norm": 1.5486682653427124, |
| "learning_rate": 2.0936739659367396e-05, |
| "loss": 0.7897, |
| "step": 2400 |
| }, |
| { |
| "epoch": 5.840438489646772, |
| "eval_loss": 0.6454324126243591, |
| "eval_runtime": 207.86, |
| "eval_samples_per_second": 7.924, |
| "eval_steps_per_second": 0.991, |
| "eval_wer": 0.5443674894926893, |
| "step": 2400 |
| }, |
| { |
| "epoch": 6.082825822168088, |
| "grad_norm": 0.9000282287597656, |
| "learning_rate": 1.972019464720195e-05, |
| "loss": 0.7723, |
| "step": 2500 |
| }, |
| { |
| "epoch": 6.082825822168088, |
| "eval_loss": 0.6421072483062744, |
| "eval_runtime": 208.0417, |
| "eval_samples_per_second": 7.917, |
| "eval_steps_per_second": 0.99, |
| "eval_wer": 0.5446042739596283, |
| "step": 2500 |
| }, |
| { |
| "epoch": 6.326431181485993, |
| "grad_norm": 1.1051301956176758, |
| "learning_rate": 1.8515815085158152e-05, |
| "loss": 0.7805, |
| "step": 2600 |
| }, |
| { |
| "epoch": 6.326431181485993, |
| "eval_loss": 0.6392606496810913, |
| "eval_runtime": 207.7887, |
| "eval_samples_per_second": 7.926, |
| "eval_steps_per_second": 0.991, |
| "eval_wer": 0.5412892914224827, |
| "step": 2600 |
| }, |
| { |
| "epoch": 6.570036540803898, |
| "grad_norm": 0.7573370337486267, |
| "learning_rate": 1.7299270072992702e-05, |
| "loss": 0.7974, |
| "step": 2700 |
| }, |
| { |
| "epoch": 6.570036540803898, |
| "eval_loss": 0.6364970803260803, |
| "eval_runtime": 207.756, |
| "eval_samples_per_second": 7.928, |
| "eval_steps_per_second": 0.992, |
| "eval_wer": 0.5396318001539099, |
| "step": 2700 |
| }, |
| { |
| "epoch": 6.8136419001218025, |
| "grad_norm": 1.2818819284439087, |
| "learning_rate": 1.608272506082725e-05, |
| "loss": 0.7794, |
| "step": 2800 |
| }, |
| { |
| "epoch": 6.8136419001218025, |
| "eval_loss": 0.6344229578971863, |
| "eval_runtime": 207.1477, |
| "eval_samples_per_second": 7.951, |
| "eval_steps_per_second": 0.994, |
| "eval_wer": 0.5391582312200319, |
| "step": 2800 |
| }, |
| { |
| "epoch": 7.056029232643118, |
| "grad_norm": 1.371817708015442, |
| "learning_rate": 1.4866180048661801e-05, |
| "loss": 0.7676, |
| "step": 2900 |
| }, |
| { |
| "epoch": 7.056029232643118, |
| "eval_loss": 0.6325829029083252, |
| "eval_runtime": 207.5164, |
| "eval_samples_per_second": 7.937, |
| "eval_steps_per_second": 0.993, |
| "eval_wer": 0.5388622506363583, |
| "step": 2900 |
| }, |
| { |
| "epoch": 7.2996345919610235, |
| "grad_norm": 0.6480887532234192, |
| "learning_rate": 1.364963503649635e-05, |
| "loss": 0.7627, |
| "step": 3000 |
| }, |
| { |
| "epoch": 7.2996345919610235, |
| "eval_loss": 0.6304593682289124, |
| "eval_runtime": 206.7906, |
| "eval_samples_per_second": 7.965, |
| "eval_steps_per_second": 0.996, |
| "eval_wer": 0.5392766234535015, |
| "step": 3000 |
| }, |
| { |
| "epoch": 7.543239951278928, |
| "grad_norm": 0.8857569694519043, |
| "learning_rate": 1.2445255474452555e-05, |
| "loss": 0.7881, |
| "step": 3100 |
| }, |
| { |
| "epoch": 7.543239951278928, |
| "eval_loss": 0.6281934976577759, |
| "eval_runtime": 207.7645, |
| "eval_samples_per_second": 7.927, |
| "eval_steps_per_second": 0.992, |
| "eval_wer": 0.5379151127686024, |
| "step": 3100 |
| }, |
| { |
| "epoch": 7.786845310596833, |
| "grad_norm": 3.5432286262512207, |
| "learning_rate": 1.1228710462287105e-05, |
| "loss": 0.7689, |
| "step": 3200 |
| }, |
| { |
| "epoch": 7.786845310596833, |
| "eval_loss": 0.6266854405403137, |
| "eval_runtime": 209.1578, |
| "eval_samples_per_second": 7.874, |
| "eval_steps_per_second": 0.985, |
| "eval_wer": 0.5341857574143136, |
| "step": 3200 |
| }, |
| { |
| "epoch": 8.02923264311815, |
| "grad_norm": 0.5050083994865417, |
| "learning_rate": 1.0012165450121655e-05, |
| "loss": 0.7784, |
| "step": 3300 |
| }, |
| { |
| "epoch": 8.02923264311815, |
| "eval_loss": 0.6252668499946594, |
| "eval_runtime": 208.1091, |
| "eval_samples_per_second": 7.914, |
| "eval_steps_per_second": 0.99, |
| "eval_wer": 0.5369679749008465, |
| "step": 3300 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 4110, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 10, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4.631639647448512e+19, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
