| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.9193857965451055, |
| "eval_steps": 200, |
| "global_step": 3000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.03198976327575176, |
| "grad_norm": 0.27879878878593445, |
| "learning_rate": 0.0009895500106632544, |
| "loss": 2.0325, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.06397952655150352, |
| "grad_norm": 0.30282971262931824, |
| "learning_rate": 0.0009788867562380039, |
| "loss": 1.9529, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.09596928982725528, |
| "grad_norm": 0.3248352110385895, |
| "learning_rate": 0.0009682235018127533, |
| "loss": 2.0104, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.12795905310300704, |
| "grad_norm": 0.3304312527179718, |
| "learning_rate": 0.0009575602473875027, |
| "loss": 1.9798, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.12795905310300704, |
| "eval_loss": 1.6201289892196655, |
| "eval_rouge1": 24.9524, |
| "eval_rouge2": 11.8115, |
| "eval_rougeL": 20.4228, |
| "eval_rougeLsum": 23.4167, |
| "eval_runtime": 315.9771, |
| "eval_samples_per_second": 7.912, |
| "eval_steps_per_second": 0.497, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.1599488163787588, |
| "grad_norm": 0.35063809156417847, |
| "learning_rate": 0.000946896992962252, |
| "loss": 2.0127, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.19193857965451055, |
| "grad_norm": 0.3319188356399536, |
| "learning_rate": 0.0009362337385370016, |
| "loss": 1.9046, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.22392834293026231, |
| "grad_norm": 0.35874632000923157, |
| "learning_rate": 0.0009255704841117509, |
| "loss": 1.983, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.2559181062060141, |
| "grad_norm": 0.37553051114082336, |
| "learning_rate": 0.0009149072296865004, |
| "loss": 1.9382, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.2559181062060141, |
| "eval_loss": 1.6310702562332153, |
| "eval_rouge1": 24.8892, |
| "eval_rouge2": 11.7408, |
| "eval_rougeL": 20.3925, |
| "eval_rougeLsum": 23.3202, |
| "eval_runtime": 315.7526, |
| "eval_samples_per_second": 7.918, |
| "eval_steps_per_second": 0.497, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.28790786948176583, |
| "grad_norm": 0.3704718053340912, |
| "learning_rate": 0.0009042439752612498, |
| "loss": 1.975, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.3198976327575176, |
| "grad_norm": 0.33435943722724915, |
| "learning_rate": 0.0008935807208359991, |
| "loss": 1.9376, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.35188739603326935, |
| "grad_norm": 0.3917178213596344, |
| "learning_rate": 0.0008829174664107486, |
| "loss": 1.9519, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.3838771593090211, |
| "grad_norm": 0.4132656455039978, |
| "learning_rate": 0.000872254211985498, |
| "loss": 1.9487, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.3838771593090211, |
| "eval_loss": 1.62241792678833, |
| "eval_rouge1": 24.8193, |
| "eval_rouge2": 11.7477, |
| "eval_rougeL": 20.3401, |
| "eval_rougeLsum": 23.362, |
| "eval_runtime": 315.9731, |
| "eval_samples_per_second": 7.912, |
| "eval_steps_per_second": 0.497, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.41586692258477287, |
| "grad_norm": 0.3778621554374695, |
| "learning_rate": 0.0008615909575602474, |
| "loss": 1.9853, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.44785668586052463, |
| "grad_norm": 0.35340753197669983, |
| "learning_rate": 0.0008509277031349968, |
| "loss": 1.9293, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.4798464491362764, |
| "grad_norm": 0.3727020025253296, |
| "learning_rate": 0.0008402644487097463, |
| "loss": 1.9616, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.5118362124120281, |
| "grad_norm": 0.3767358958721161, |
| "learning_rate": 0.0008296011942844956, |
| "loss": 1.9872, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.5118362124120281, |
| "eval_loss": 1.6222777366638184, |
| "eval_rouge1": 24.8043, |
| "eval_rouge2": 11.6712, |
| "eval_rougeL": 20.2708, |
| "eval_rougeLsum": 23.3073, |
| "eval_runtime": 316.4732, |
| "eval_samples_per_second": 7.9, |
| "eval_steps_per_second": 0.496, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.5438259756877799, |
| "grad_norm": 0.36367788910865784, |
| "learning_rate": 0.0008189379398592451, |
| "loss": 1.94, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.5758157389635317, |
| "grad_norm": 0.3763476610183716, |
| "learning_rate": 0.0008082746854339945, |
| "loss": 1.9334, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.6078055022392834, |
| "grad_norm": 0.3726671040058136, |
| "learning_rate": 0.0007976114310087439, |
| "loss": 1.9465, |
| "step": 950 |
| }, |
| { |
| "epoch": 0.6397952655150352, |
| "grad_norm": 0.43825629353523254, |
| "learning_rate": 0.0007869481765834933, |
| "loss": 1.9892, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.6397952655150352, |
| "eval_loss": 1.6152864694595337, |
| "eval_rouge1": 25.0181, |
| "eval_rouge2": 11.7926, |
| "eval_rougeL": 20.4384, |
| "eval_rougeLsum": 23.5256, |
| "eval_runtime": 315.8831, |
| "eval_samples_per_second": 7.914, |
| "eval_steps_per_second": 0.497, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.6717850287907869, |
| "grad_norm": 0.3792516589164734, |
| "learning_rate": 0.0007762849221582428, |
| "loss": 1.9282, |
| "step": 1050 |
| }, |
| { |
| "epoch": 0.7037747920665387, |
| "grad_norm": 0.3944825530052185, |
| "learning_rate": 0.0007656216677329921, |
| "loss": 1.9837, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.7357645553422905, |
| "grad_norm": 0.35828277468681335, |
| "learning_rate": 0.0007549584133077415, |
| "loss": 1.9838, |
| "step": 1150 |
| }, |
| { |
| "epoch": 0.7677543186180422, |
| "grad_norm": 0.36714279651641846, |
| "learning_rate": 0.000744295158882491, |
| "loss": 1.9239, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.7677543186180422, |
| "eval_loss": 1.6160937547683716, |
| "eval_rouge1": 25.0574, |
| "eval_rouge2": 11.7988, |
| "eval_rougeL": 20.5007, |
| "eval_rougeLsum": 23.6136, |
| "eval_runtime": 315.6243, |
| "eval_samples_per_second": 7.921, |
| "eval_steps_per_second": 0.497, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.799744081893794, |
| "grad_norm": 0.42614004015922546, |
| "learning_rate": 0.0007336319044572403, |
| "loss": 1.9428, |
| "step": 1250 |
| }, |
| { |
| "epoch": 0.8317338451695457, |
| "grad_norm": 0.3988937735557556, |
| "learning_rate": 0.0007229686500319898, |
| "loss": 1.9007, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.8637236084452975, |
| "grad_norm": 0.38287872076034546, |
| "learning_rate": 0.0007123053956067392, |
| "loss": 1.9454, |
| "step": 1350 |
| }, |
| { |
| "epoch": 0.8957133717210493, |
| "grad_norm": 0.35704633593559265, |
| "learning_rate": 0.0007016421411814886, |
| "loss": 1.9169, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.8957133717210493, |
| "eval_loss": 1.6169191598892212, |
| "eval_rouge1": 24.9579, |
| "eval_rouge2": 11.8434, |
| "eval_rougeL": 20.4622, |
| "eval_rougeLsum": 23.4492, |
| "eval_runtime": 315.5353, |
| "eval_samples_per_second": 7.923, |
| "eval_steps_per_second": 0.498, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.927703134996801, |
| "grad_norm": 0.38228747248649597, |
| "learning_rate": 0.000690978886756238, |
| "loss": 1.937, |
| "step": 1450 |
| }, |
| { |
| "epoch": 0.9596928982725528, |
| "grad_norm": 0.4126349985599518, |
| "learning_rate": 0.0006803156323309875, |
| "loss": 1.9719, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.9916826615483045, |
| "grad_norm": 0.3939051926136017, |
| "learning_rate": 0.0006696523779057368, |
| "loss": 1.979, |
| "step": 1550 |
| }, |
| { |
| "epoch": 1.0236724248240563, |
| "grad_norm": 0.37081676721572876, |
| "learning_rate": 0.0006589891234804863, |
| "loss": 1.9215, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.0236724248240563, |
| "eval_loss": 1.6208561658859253, |
| "eval_rouge1": 24.8442, |
| "eval_rouge2": 11.7417, |
| "eval_rougeL": 20.3416, |
| "eval_rougeLsum": 23.335, |
| "eval_runtime": 315.7021, |
| "eval_samples_per_second": 7.919, |
| "eval_steps_per_second": 0.497, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.055662188099808, |
| "grad_norm": 0.38813772797584534, |
| "learning_rate": 0.0006483258690552357, |
| "loss": 1.9858, |
| "step": 1650 |
| }, |
| { |
| "epoch": 1.0876519513755598, |
| "grad_norm": 0.3699555993080139, |
| "learning_rate": 0.0006376626146299851, |
| "loss": 1.928, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.1196417146513116, |
| "grad_norm": 0.36284947395324707, |
| "learning_rate": 0.0006269993602047345, |
| "loss": 1.9219, |
| "step": 1750 |
| }, |
| { |
| "epoch": 1.1516314779270633, |
| "grad_norm": 0.3615477383136749, |
| "learning_rate": 0.000616336105779484, |
| "loss": 1.9528, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.1516314779270633, |
| "eval_loss": 1.6172361373901367, |
| "eval_rouge1": 24.8712, |
| "eval_rouge2": 11.7236, |
| "eval_rougeL": 20.3555, |
| "eval_rougeLsum": 23.3763, |
| "eval_runtime": 316.1677, |
| "eval_samples_per_second": 7.907, |
| "eval_steps_per_second": 0.497, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.183621241202815, |
| "grad_norm": 0.40732404589653015, |
| "learning_rate": 0.0006056728513542333, |
| "loss": 1.8972, |
| "step": 1850 |
| }, |
| { |
| "epoch": 1.2156110044785668, |
| "grad_norm": 0.3976059556007385, |
| "learning_rate": 0.0005950095969289827, |
| "loss": 1.9675, |
| "step": 1900 |
| }, |
| { |
| "epoch": 1.2476007677543186, |
| "grad_norm": 0.3994428217411041, |
| "learning_rate": 0.0005843463425037322, |
| "loss": 1.8981, |
| "step": 1950 |
| }, |
| { |
| "epoch": 1.2795905310300704, |
| "grad_norm": 0.434885174036026, |
| "learning_rate": 0.0005736830880784815, |
| "loss": 1.9293, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.2795905310300704, |
| "eval_loss": 1.6173455715179443, |
| "eval_rouge1": 24.913, |
| "eval_rouge2": 11.7915, |
| "eval_rougeL": 20.3985, |
| "eval_rougeLsum": 23.4071, |
| "eval_runtime": 315.9307, |
| "eval_samples_per_second": 7.913, |
| "eval_steps_per_second": 0.497, |
| "step": 2000 |
| }, |
| { |
| "epoch": 1.3115802943058221, |
| "grad_norm": 0.407560795545578, |
| "learning_rate": 0.000563019833653231, |
| "loss": 1.9515, |
| "step": 2050 |
| }, |
| { |
| "epoch": 1.3435700575815739, |
| "grad_norm": 0.4360503554344177, |
| "learning_rate": 0.0005523565792279804, |
| "loss": 1.9395, |
| "step": 2100 |
| }, |
| { |
| "epoch": 1.3755598208573256, |
| "grad_norm": 0.3891420364379883, |
| "learning_rate": 0.0005416933248027298, |
| "loss": 1.9251, |
| "step": 2150 |
| }, |
| { |
| "epoch": 1.4075495841330774, |
| "grad_norm": 0.40385594964027405, |
| "learning_rate": 0.0005310300703774792, |
| "loss": 1.908, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.4075495841330774, |
| "eval_loss": 1.6220018863677979, |
| "eval_rouge1": 24.9961, |
| "eval_rouge2": 11.8233, |
| "eval_rougeL": 20.4102, |
| "eval_rougeLsum": 23.4576, |
| "eval_runtime": 315.458, |
| "eval_samples_per_second": 7.925, |
| "eval_steps_per_second": 0.498, |
| "step": 2200 |
| }, |
| { |
| "epoch": 1.4395393474088292, |
| "grad_norm": 0.39512693881988525, |
| "learning_rate": 0.0005203668159522287, |
| "loss": 1.9284, |
| "step": 2250 |
| }, |
| { |
| "epoch": 1.471529110684581, |
| "grad_norm": 0.3453839123249054, |
| "learning_rate": 0.000509703561526978, |
| "loss": 1.9213, |
| "step": 2300 |
| }, |
| { |
| "epoch": 1.5035188739603327, |
| "grad_norm": 0.37096330523490906, |
| "learning_rate": 0.0004990403071017275, |
| "loss": 1.9006, |
| "step": 2350 |
| }, |
| { |
| "epoch": 1.5355086372360844, |
| "grad_norm": 0.403342068195343, |
| "learning_rate": 0.0004883770526764769, |
| "loss": 1.9156, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.5355086372360844, |
| "eval_loss": 1.6193974018096924, |
| "eval_rouge1": 24.831, |
| "eval_rouge2": 11.6708, |
| "eval_rougeL": 20.2876, |
| "eval_rougeLsum": 23.2907, |
| "eval_runtime": 315.8804, |
| "eval_samples_per_second": 7.914, |
| "eval_steps_per_second": 0.497, |
| "step": 2400 |
| }, |
| { |
| "epoch": 1.5674984005118362, |
| "grad_norm": 0.39487147331237793, |
| "learning_rate": 0.0004777137982512263, |
| "loss": 1.9342, |
| "step": 2450 |
| }, |
| { |
| "epoch": 1.599488163787588, |
| "grad_norm": 0.38196542859077454, |
| "learning_rate": 0.0004670505438259757, |
| "loss": 1.9135, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.6314779270633397, |
| "grad_norm": 0.37945207953453064, |
| "learning_rate": 0.0004563872894007251, |
| "loss": 1.8889, |
| "step": 2550 |
| }, |
| { |
| "epoch": 1.6634676903390915, |
| "grad_norm": 0.4320591688156128, |
| "learning_rate": 0.00044572403497547455, |
| "loss": 1.8561, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.6634676903390915, |
| "eval_loss": 1.6163318157196045, |
| "eval_rouge1": 24.955, |
| "eval_rouge2": 11.8385, |
| "eval_rougeL": 20.3963, |
| "eval_rougeLsum": 23.4865, |
| "eval_runtime": 316.1324, |
| "eval_samples_per_second": 7.908, |
| "eval_steps_per_second": 0.497, |
| "step": 2600 |
| }, |
| { |
| "epoch": 1.6954574536148432, |
| "grad_norm": 0.3730495572090149, |
| "learning_rate": 0.00043506078055022393, |
| "loss": 1.9485, |
| "step": 2650 |
| }, |
| { |
| "epoch": 1.727447216890595, |
| "grad_norm": 0.36828505992889404, |
| "learning_rate": 0.00042439752612497336, |
| "loss": 1.9225, |
| "step": 2700 |
| }, |
| { |
| "epoch": 1.7594369801663468, |
| "grad_norm": 0.3844313323497772, |
| "learning_rate": 0.0004137342716997228, |
| "loss": 1.9328, |
| "step": 2750 |
| }, |
| { |
| "epoch": 1.7914267434420985, |
| "grad_norm": 0.36565858125686646, |
| "learning_rate": 0.0004030710172744721, |
| "loss": 1.9838, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.7914267434420985, |
| "eval_loss": 1.609842300415039, |
| "eval_rouge1": 25.0754, |
| "eval_rouge2": 11.842, |
| "eval_rougeL": 20.5251, |
| "eval_rougeLsum": 23.6108, |
| "eval_runtime": 315.76, |
| "eval_samples_per_second": 7.917, |
| "eval_steps_per_second": 0.497, |
| "step": 2800 |
| }, |
| { |
| "epoch": 1.8234165067178503, |
| "grad_norm": 0.3444318473339081, |
| "learning_rate": 0.00039240776284922156, |
| "loss": 1.9054, |
| "step": 2850 |
| }, |
| { |
| "epoch": 1.855406269993602, |
| "grad_norm": 0.3635132610797882, |
| "learning_rate": 0.000381744508423971, |
| "loss": 1.9056, |
| "step": 2900 |
| }, |
| { |
| "epoch": 1.8873960332693538, |
| "grad_norm": 0.392283171415329, |
| "learning_rate": 0.0003710812539987204, |
| "loss": 1.9451, |
| "step": 2950 |
| }, |
| { |
| "epoch": 1.9193857965451055, |
| "grad_norm": 0.36240875720977783, |
| "learning_rate": 0.0003604179995734698, |
| "loss": 1.9532, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.9193857965451055, |
| "eval_loss": 1.6145464181900024, |
| "eval_rouge1": 25.1424, |
| "eval_rouge2": 12.0047, |
| "eval_rougeL": 20.6482, |
| "eval_rougeLsum": 23.6427, |
| "eval_runtime": 315.6824, |
| "eval_samples_per_second": 7.919, |
| "eval_steps_per_second": 0.497, |
| "step": 3000 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 4689, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 200, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.299331274702848e+16, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
