| { | |
| "best_global_step": 10000, | |
| "best_metric": 7.6356892585754395, | |
| "best_model_checkpoint": "bert_base_rand_5_v1/checkpoint-10000", | |
| "epoch": 25.0, | |
| "eval_steps": 10000, | |
| "global_step": 59550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.20990764063811923, | |
| "grad_norm": 2.498877763748169, | |
| "learning_rate": 5e-06, | |
| "loss": 10.8092, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41981528127623846, | |
| "grad_norm": 2.6819911003112793, | |
| "learning_rate": 1e-05, | |
| "loss": 9.1411, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6297229219143576, | |
| "grad_norm": 1.7457034587860107, | |
| "learning_rate": 1.5e-05, | |
| "loss": 8.596, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8396305625524769, | |
| "grad_norm": 1.8576099872589111, | |
| "learning_rate": 2e-05, | |
| "loss": 8.4357, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0495382031905962, | |
| "grad_norm": 1.5093258619308472, | |
| "learning_rate": 2.5e-05, | |
| "loss": 8.3197, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.2594458438287153, | |
| "grad_norm": 1.7299160957336426, | |
| "learning_rate": 3e-05, | |
| "loss": 8.2113, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4693534844668346, | |
| "grad_norm": 1.711667537689209, | |
| "learning_rate": 3.5e-05, | |
| "loss": 8.1224, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.6792611251049538, | |
| "grad_norm": 2.2058727741241455, | |
| "learning_rate": 4e-05, | |
| "loss": 8.0566, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.8891687657430731, | |
| "grad_norm": 1.0799951553344727, | |
| "learning_rate": 4.5e-05, | |
| "loss": 8.0003, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0990764063811924, | |
| "grad_norm": 1.8451857566833496, | |
| "learning_rate": 5e-05, | |
| "loss": 7.9515, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.3089840470193117, | |
| "grad_norm": 1.6103696823120117, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 7.9105, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.5188916876574305, | |
| "grad_norm": 1.6846959590911865, | |
| "learning_rate": 6e-05, | |
| "loss": 7.8761, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.72879932829555, | |
| "grad_norm": 1.9888107776641846, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 7.8423, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.938706968933669, | |
| "grad_norm": 2.885254144668579, | |
| "learning_rate": 7e-05, | |
| "loss": 7.813, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.1486146095717884, | |
| "grad_norm": 3.383962631225586, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 7.7895, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.3585222502099077, | |
| "grad_norm": 2.675636053085327, | |
| "learning_rate": 8e-05, | |
| "loss": 7.7668, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.568429890848027, | |
| "grad_norm": 1.3491432666778564, | |
| "learning_rate": 8.5e-05, | |
| "loss": 7.7513, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.7783375314861463, | |
| "grad_norm": 1.60332190990448, | |
| "learning_rate": 9e-05, | |
| "loss": 7.7253, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.988245172124265, | |
| "grad_norm": 3.6063122749328613, | |
| "learning_rate": 9.5e-05, | |
| "loss": 7.7062, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.198152812762385, | |
| "grad_norm": 2.8932175636291504, | |
| "learning_rate": 0.0001, | |
| "loss": 7.6842, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.198152812762385, | |
| "eval_accuracy": 0.15100587105929492, | |
| "eval_loss": 7.6356892585754395, | |
| "eval_runtime": 1.0609, | |
| "eval_samples_per_second": 451.522, | |
| "eval_steps_per_second": 4.713, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.408060453400504, | |
| "grad_norm": 2.1227481365203857, | |
| "learning_rate": 9.899091826437942e-05, | |
| "loss": 7.6683, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.617968094038623, | |
| "grad_norm": 3.855611562728882, | |
| "learning_rate": 9.798183652875883e-05, | |
| "loss": 7.6508, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.827875734676742, | |
| "grad_norm": 2.028477191925049, | |
| "learning_rate": 9.697275479313825e-05, | |
| "loss": 7.6425, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.037783375314861, | |
| "grad_norm": 1.9752072095870972, | |
| "learning_rate": 9.596367305751767e-05, | |
| "loss": 7.623, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.247691015952981, | |
| "grad_norm": 2.0890986919403076, | |
| "learning_rate": 9.495459132189707e-05, | |
| "loss": 7.5849, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.4575986565911, | |
| "grad_norm": 3.943803071975708, | |
| "learning_rate": 9.394550958627649e-05, | |
| "loss": 7.5752, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.667506297229219, | |
| "grad_norm": 1.2731233835220337, | |
| "learning_rate": 9.293642785065591e-05, | |
| "loss": 7.5748, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 5.877413937867338, | |
| "grad_norm": 1.713611364364624, | |
| "learning_rate": 9.192734611503532e-05, | |
| "loss": 7.5662, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.087321578505458, | |
| "grad_norm": 2.786539316177368, | |
| "learning_rate": 9.091826437941474e-05, | |
| "loss": 7.5349, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 6.297229219143577, | |
| "grad_norm": 2.8720133304595947, | |
| "learning_rate": 8.990918264379415e-05, | |
| "loss": 7.5123, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.507136859781696, | |
| "grad_norm": 2.2100372314453125, | |
| "learning_rate": 8.890010090817356e-05, | |
| "loss": 7.5056, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 6.717044500419815, | |
| "grad_norm": 2.491044521331787, | |
| "learning_rate": 8.789101917255298e-05, | |
| "loss": 7.5109, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 6.926952141057934, | |
| "grad_norm": 1.7924177646636963, | |
| "learning_rate": 8.68819374369324e-05, | |
| "loss": 7.5064, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 7.136859781696054, | |
| "grad_norm": 3.66336727142334, | |
| "learning_rate": 8.58728557013118e-05, | |
| "loss": 7.4614, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 7.346767422334173, | |
| "grad_norm": 2.913663148880005, | |
| "learning_rate": 8.486377396569122e-05, | |
| "loss": 7.4427, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 7.5566750629722925, | |
| "grad_norm": 2.750014305114746, | |
| "learning_rate": 8.385469223007064e-05, | |
| "loss": 7.4495, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 7.766582703610411, | |
| "grad_norm": 2.7463138103485107, | |
| "learning_rate": 8.284561049445006e-05, | |
| "loss": 7.455, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 7.97649034424853, | |
| "grad_norm": 3.206331729888916, | |
| "learning_rate": 8.183652875882947e-05, | |
| "loss": 7.4559, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 8.18639798488665, | |
| "grad_norm": 2.5946578979492188, | |
| "learning_rate": 8.082744702320889e-05, | |
| "loss": 7.3741, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 8.39630562552477, | |
| "grad_norm": 3.3631999492645264, | |
| "learning_rate": 7.98183652875883e-05, | |
| "loss": 7.3809, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.39630562552477, | |
| "eval_accuracy": 0.15272594629478942, | |
| "eval_loss": 7.655212879180908, | |
| "eval_runtime": 1.0783, | |
| "eval_samples_per_second": 444.222, | |
| "eval_steps_per_second": 4.637, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.606213266162888, | |
| "grad_norm": 1.9286588430404663, | |
| "learning_rate": 7.880928355196771e-05, | |
| "loss": 7.3817, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 8.816120906801007, | |
| "grad_norm": 2.3397583961486816, | |
| "learning_rate": 7.780020181634713e-05, | |
| "loss": 7.3931, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 9.026028547439127, | |
| "grad_norm": 4.4994282722473145, | |
| "learning_rate": 7.679112008072655e-05, | |
| "loss": 7.3719, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 9.235936188077247, | |
| "grad_norm": 3.9538917541503906, | |
| "learning_rate": 7.578203834510595e-05, | |
| "loss": 7.255, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 9.445843828715365, | |
| "grad_norm": 3.1948907375335693, | |
| "learning_rate": 7.477295660948537e-05, | |
| "loss": 7.2712, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 9.655751469353484, | |
| "grad_norm": 4.393702507019043, | |
| "learning_rate": 7.376387487386479e-05, | |
| "loss": 7.2824, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 9.865659109991604, | |
| "grad_norm": 3.8742611408233643, | |
| "learning_rate": 7.27547931382442e-05, | |
| "loss": 7.2823, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 10.075566750629722, | |
| "grad_norm": 4.709822177886963, | |
| "learning_rate": 7.174571140262362e-05, | |
| "loss": 7.2037, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 10.285474391267842, | |
| "grad_norm": 5.138158321380615, | |
| "learning_rate": 7.073662966700304e-05, | |
| "loss": 7.0851, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 10.495382031905962, | |
| "grad_norm": 4.4400458335876465, | |
| "learning_rate": 6.972754793138244e-05, | |
| "loss": 7.0995, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 10.705289672544081, | |
| "grad_norm": 5.382376670837402, | |
| "learning_rate": 6.871846619576186e-05, | |
| "loss": 7.1128, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 10.9151973131822, | |
| "grad_norm": 3.9833223819732666, | |
| "learning_rate": 6.770938446014128e-05, | |
| "loss": 7.1245, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 11.125104953820319, | |
| "grad_norm": 4.84190034866333, | |
| "learning_rate": 6.670030272452068e-05, | |
| "loss": 6.9818, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 11.335012594458439, | |
| "grad_norm": 6.207947731018066, | |
| "learning_rate": 6.56912209889001e-05, | |
| "loss": 6.9052, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 11.544920235096557, | |
| "grad_norm": 4.902656555175781, | |
| "learning_rate": 6.468213925327952e-05, | |
| "loss": 6.9296, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 11.754827875734676, | |
| "grad_norm": 4.875358581542969, | |
| "learning_rate": 6.367305751765893e-05, | |
| "loss": 6.9338, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 11.964735516372796, | |
| "grad_norm": 4.797720432281494, | |
| "learning_rate": 6.266397578203835e-05, | |
| "loss": 6.9559, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 12.174643157010916, | |
| "grad_norm": 5.911436557769775, | |
| "learning_rate": 6.165489404641777e-05, | |
| "loss": 6.7805, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 12.384550797649034, | |
| "grad_norm": 4.664194107055664, | |
| "learning_rate": 6.064581231079718e-05, | |
| "loss": 6.7606, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 12.594458438287154, | |
| "grad_norm": 6.650626182556152, | |
| "learning_rate": 5.963673057517659e-05, | |
| "loss": 6.7846, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.594458438287154, | |
| "eval_accuracy": 0.15490947757226403, | |
| "eval_loss": 8.528319358825684, | |
| "eval_runtime": 1.0223, | |
| "eval_samples_per_second": 468.535, | |
| "eval_steps_per_second": 4.891, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.804366078925273, | |
| "grad_norm": 7.026340961456299, | |
| "learning_rate": 5.8627648839556e-05, | |
| "loss": 6.7948, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 13.014273719563391, | |
| "grad_norm": 7.709275245666504, | |
| "learning_rate": 5.7618567103935416e-05, | |
| "loss": 6.7931, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 13.224181360201511, | |
| "grad_norm": 4.2592549324035645, | |
| "learning_rate": 5.660948536831484e-05, | |
| "loss": 6.6292, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 13.43408900083963, | |
| "grad_norm": 5.644862174987793, | |
| "learning_rate": 5.5600403632694254e-05, | |
| "loss": 6.6466, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 13.64399664147775, | |
| "grad_norm": 6.733758449554443, | |
| "learning_rate": 5.4591321897073666e-05, | |
| "loss": 6.6716, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 13.853904282115868, | |
| "grad_norm": 6.484296798706055, | |
| "learning_rate": 5.358224016145308e-05, | |
| "loss": 6.6785, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 14.063811922753988, | |
| "grad_norm": 3.7475457191467285, | |
| "learning_rate": 5.257315842583249e-05, | |
| "loss": 6.6305, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 14.273719563392108, | |
| "grad_norm": 4.5703959465026855, | |
| "learning_rate": 5.15640766902119e-05, | |
| "loss": 6.5417, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 14.483627204030226, | |
| "grad_norm": 7.1667656898498535, | |
| "learning_rate": 5.055499495459133e-05, | |
| "loss": 6.5533, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 14.693534844668346, | |
| "grad_norm": 6.3863630294799805, | |
| "learning_rate": 4.954591321897074e-05, | |
| "loss": 6.5615, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 14.903442485306465, | |
| "grad_norm": 5.614595413208008, | |
| "learning_rate": 4.853683148335015e-05, | |
| "loss": 6.5759, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 15.113350125944585, | |
| "grad_norm": 6.775130271911621, | |
| "learning_rate": 4.7527749747729566e-05, | |
| "loss": 6.5079, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 15.323257766582703, | |
| "grad_norm": 5.867656230926514, | |
| "learning_rate": 4.6518668012108985e-05, | |
| "loss": 6.4581, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 15.533165407220823, | |
| "grad_norm": 6.446193695068359, | |
| "learning_rate": 4.55095862764884e-05, | |
| "loss": 6.4621, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 15.743073047858942, | |
| "grad_norm": 5.2496113777160645, | |
| "learning_rate": 4.450050454086781e-05, | |
| "loss": 6.4764, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 15.95298068849706, | |
| "grad_norm": 7.304086685180664, | |
| "learning_rate": 4.349142280524723e-05, | |
| "loss": 6.4834, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 16.162888329135182, | |
| "grad_norm": 6.148995399475098, | |
| "learning_rate": 4.248234106962664e-05, | |
| "loss": 6.4038, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 16.3727959697733, | |
| "grad_norm": 7.440193176269531, | |
| "learning_rate": 4.147325933400605e-05, | |
| "loss": 6.3928, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 16.582703610411418, | |
| "grad_norm": 5.378032207489014, | |
| "learning_rate": 4.046417759838547e-05, | |
| "loss": 6.4001, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 16.79261125104954, | |
| "grad_norm": 7.020175933837891, | |
| "learning_rate": 3.9455095862764884e-05, | |
| "loss": 6.3984, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 16.79261125104954, | |
| "eval_accuracy": 0.15204104519457218, | |
| "eval_loss": 9.142778396606445, | |
| "eval_runtime": 1.0058, | |
| "eval_samples_per_second": 476.214, | |
| "eval_steps_per_second": 4.971, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 17.002518891687657, | |
| "grad_norm": 7.505725860595703, | |
| "learning_rate": 3.84460141271443e-05, | |
| "loss": 6.4088, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 17.212426532325775, | |
| "grad_norm": 6.304658889770508, | |
| "learning_rate": 3.7436932391523715e-05, | |
| "loss": 6.3232, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 17.422334172963897, | |
| "grad_norm": 5.694614410400391, | |
| "learning_rate": 3.642785065590313e-05, | |
| "loss": 6.3307, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 17.632241813602015, | |
| "grad_norm": 8.763103485107422, | |
| "learning_rate": 3.541876892028255e-05, | |
| "loss": 6.3344, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 17.842149454240133, | |
| "grad_norm": 6.419318199157715, | |
| "learning_rate": 3.440968718466196e-05, | |
| "loss": 6.3445, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 18.052057094878254, | |
| "grad_norm": 7.152398586273193, | |
| "learning_rate": 3.340060544904137e-05, | |
| "loss": 6.3267, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 18.261964735516372, | |
| "grad_norm": 5.1622514724731445, | |
| "learning_rate": 3.239152371342079e-05, | |
| "loss": 6.2759, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 18.471872376154494, | |
| "grad_norm": 5.933616638183594, | |
| "learning_rate": 3.13824419778002e-05, | |
| "loss": 6.282, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 18.68178001679261, | |
| "grad_norm": 6.564027309417725, | |
| "learning_rate": 3.0373360242179615e-05, | |
| "loss": 6.2925, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 18.89168765743073, | |
| "grad_norm": 5.797423362731934, | |
| "learning_rate": 2.9364278506559034e-05, | |
| "loss": 6.2952, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 19.10159529806885, | |
| "grad_norm": 5.061058521270752, | |
| "learning_rate": 2.8355196770938446e-05, | |
| "loss": 6.2541, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 19.31150293870697, | |
| "grad_norm": 4.466089248657227, | |
| "learning_rate": 2.734611503531786e-05, | |
| "loss": 6.2238, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 19.521410579345087, | |
| "grad_norm": 5.071774482727051, | |
| "learning_rate": 2.6337033299697278e-05, | |
| "loss": 6.2312, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 19.73131821998321, | |
| "grad_norm": 6.869780540466309, | |
| "learning_rate": 2.532795156407669e-05, | |
| "loss": 6.2324, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 19.941225860621326, | |
| "grad_norm": 6.5982770919799805, | |
| "learning_rate": 2.4318869828456106e-05, | |
| "loss": 6.2281, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 20.151133501259444, | |
| "grad_norm": 6.127167701721191, | |
| "learning_rate": 2.330978809283552e-05, | |
| "loss": 6.1875, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 20.361041141897566, | |
| "grad_norm": 4.664445877075195, | |
| "learning_rate": 2.2300706357214934e-05, | |
| "loss": 6.1696, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 20.570948782535684, | |
| "grad_norm": 3.9737088680267334, | |
| "learning_rate": 2.129162462159435e-05, | |
| "loss": 6.1694, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 20.7808564231738, | |
| "grad_norm": 5.7637858390808105, | |
| "learning_rate": 2.0282542885973765e-05, | |
| "loss": 6.1707, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 20.990764063811923, | |
| "grad_norm": 2.451368808746338, | |
| "learning_rate": 1.927346115035318e-05, | |
| "loss": 6.1745, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 20.990764063811923, | |
| "eval_accuracy": 0.15190512238203382, | |
| "eval_loss": 9.834033966064453, | |
| "eval_runtime": 1.0065, | |
| "eval_samples_per_second": 475.9, | |
| "eval_steps_per_second": 4.968, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 21.20067170445004, | |
| "grad_norm": 5.885270118713379, | |
| "learning_rate": 1.8264379414732593e-05, | |
| "loss": 6.1107, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 21.410579345088163, | |
| "grad_norm": 3.8668832778930664, | |
| "learning_rate": 1.725529767911201e-05, | |
| "loss": 6.1153, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 21.62048698572628, | |
| "grad_norm": 4.112957954406738, | |
| "learning_rate": 1.6246215943491424e-05, | |
| "loss": 6.1143, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 21.8303946263644, | |
| "grad_norm": 7.522325038909912, | |
| "learning_rate": 1.5237134207870837e-05, | |
| "loss": 6.113, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 22.04030226700252, | |
| "grad_norm": 4.681398391723633, | |
| "learning_rate": 1.4228052472250252e-05, | |
| "loss": 6.1045, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 22.250209907640638, | |
| "grad_norm": 5.411729335784912, | |
| "learning_rate": 1.3218970736629668e-05, | |
| "loss": 6.0588, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 22.460117548278756, | |
| "grad_norm": 5.700425624847412, | |
| "learning_rate": 1.2209889001009082e-05, | |
| "loss": 6.0627, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 22.670025188916878, | |
| "grad_norm": 4.241090774536133, | |
| "learning_rate": 1.1200807265388496e-05, | |
| "loss": 6.0687, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 22.879932829554996, | |
| "grad_norm": 5.459275245666504, | |
| "learning_rate": 1.0191725529767912e-05, | |
| "loss": 6.0586, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 23.089840470193113, | |
| "grad_norm": 2.5863044261932373, | |
| "learning_rate": 9.182643794147326e-06, | |
| "loss": 6.0447, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 23.299748110831235, | |
| "grad_norm": 7.23863410949707, | |
| "learning_rate": 8.173562058526741e-06, | |
| "loss": 6.0258, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 23.509655751469353, | |
| "grad_norm": 2.30629301071167, | |
| "learning_rate": 7.164480322906156e-06, | |
| "loss": 6.0312, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 23.719563392107474, | |
| "grad_norm": 6.015433311462402, | |
| "learning_rate": 6.155398587285571e-06, | |
| "loss": 6.023, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 23.929471032745592, | |
| "grad_norm": 5.523674488067627, | |
| "learning_rate": 5.146316851664986e-06, | |
| "loss": 6.0248, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 24.13937867338371, | |
| "grad_norm": 4.04388427734375, | |
| "learning_rate": 4.1372351160444e-06, | |
| "loss": 6.0106, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 24.349286314021832, | |
| "grad_norm": 7.06976842880249, | |
| "learning_rate": 3.1281533804238145e-06, | |
| "loss": 6.0069, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 24.55919395465995, | |
| "grad_norm": 5.32741117477417, | |
| "learning_rate": 2.1190716448032293e-06, | |
| "loss": 6.005, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 24.769101595298068, | |
| "grad_norm": 4.379576206207275, | |
| "learning_rate": 1.109989909182644e-06, | |
| "loss": 6.0013, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 24.97900923593619, | |
| "grad_norm": 7.194871425628662, | |
| "learning_rate": 1.0090817356205853e-07, | |
| "loss": 5.9983, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "step": 59550, | |
| "total_flos": 1.5149101316039424e+18, | |
| "train_loss": 6.950234634822202, | |
| "train_runtime": 24594.6205, | |
| "train_samples_per_second": 232.408, | |
| "train_steps_per_second": 2.421 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 59550, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 25, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.5149101316039424e+18, | |
| "train_batch_size": 96, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |