| { | |
| "best_metric": 7.090852737426758, | |
| "best_model_checkpoint": "tiny_bert_km_100_v1/checkpoint-20000", | |
| "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": 2.5713536739349365, | |
| "learning_rate": 5e-06, | |
| "loss": 14.3663, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41981528127623846, | |
| "grad_norm": 1.9100619554519653, | |
| "learning_rate": 1e-05, | |
| "loss": 12.7832, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6297229219143576, | |
| "grad_norm": 2.101850986480713, | |
| "learning_rate": 1.5e-05, | |
| "loss": 11.7445, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8396305625524769, | |
| "grad_norm": 2.5743868350982666, | |
| "learning_rate": 2e-05, | |
| "loss": 11.0968, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0495382031905962, | |
| "grad_norm": 3.0924341678619385, | |
| "learning_rate": 2.5e-05, | |
| "loss": 10.0395, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.2594458438287153, | |
| "grad_norm": 3.141472816467285, | |
| "learning_rate": 3e-05, | |
| "loss": 9.1877, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4693534844668346, | |
| "grad_norm": 6.4149370193481445, | |
| "learning_rate": 3.5e-05, | |
| "loss": 8.6737, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.6792611251049538, | |
| "grad_norm": 5.3842854499816895, | |
| "learning_rate": 4e-05, | |
| "loss": 8.3257, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.8891687657430731, | |
| "grad_norm": 6.002460479736328, | |
| "learning_rate": 4.5e-05, | |
| "loss": 8.0862, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0990764063811924, | |
| "grad_norm": 4.688140392303467, | |
| "learning_rate": 5e-05, | |
| "loss": 7.8724, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.3089840470193117, | |
| "grad_norm": 6.419498920440674, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 7.7412, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.5188916876574305, | |
| "grad_norm": 8.261547088623047, | |
| "learning_rate": 6e-05, | |
| "loss": 7.6702, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.72879932829555, | |
| "grad_norm": 6.988828659057617, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 7.6109, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.938706968933669, | |
| "grad_norm": 4.17943811416626, | |
| "learning_rate": 7e-05, | |
| "loss": 7.5739, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.1486146095717884, | |
| "grad_norm": 5.65450382232666, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 7.4706, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.3585222502099077, | |
| "grad_norm": 4.269979476928711, | |
| "learning_rate": 8e-05, | |
| "loss": 7.4294, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.568429890848027, | |
| "grad_norm": 6.376979827880859, | |
| "learning_rate": 8.5e-05, | |
| "loss": 7.4368, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.7783375314861463, | |
| "grad_norm": 7.628070831298828, | |
| "learning_rate": 9e-05, | |
| "loss": 7.4121, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.988245172124265, | |
| "grad_norm": 7.433252334594727, | |
| "learning_rate": 9.5e-05, | |
| "loss": 7.3897, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.198152812762385, | |
| "grad_norm": 3.8275856971740723, | |
| "learning_rate": 0.0001, | |
| "loss": 7.2843, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.198152812762385, | |
| "eval_accuracy": 0.14722168118200285, | |
| "eval_loss": 7.432437896728516, | |
| "eval_runtime": 0.7028, | |
| "eval_samples_per_second": 681.575, | |
| "eval_steps_per_second": 7.115, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.408060453400504, | |
| "grad_norm": 5.902863502502441, | |
| "learning_rate": 9.899091826437942e-05, | |
| "loss": 7.2871, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.617968094038623, | |
| "grad_norm": 8.111076354980469, | |
| "learning_rate": 9.798183652875883e-05, | |
| "loss": 7.2748, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.827875734676742, | |
| "grad_norm": 5.7592058181762695, | |
| "learning_rate": 9.697275479313825e-05, | |
| "loss": 7.2717, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.037783375314861, | |
| "grad_norm": 6.136614799499512, | |
| "learning_rate": 9.596367305751767e-05, | |
| "loss": 7.2289, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.247691015952981, | |
| "grad_norm": 4.866406440734863, | |
| "learning_rate": 9.495459132189707e-05, | |
| "loss": 7.1308, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.4575986565911, | |
| "grad_norm": 5.56799840927124, | |
| "learning_rate": 9.394550958627649e-05, | |
| "loss": 7.1426, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.667506297229219, | |
| "grad_norm": 4.18687105178833, | |
| "learning_rate": 9.293642785065591e-05, | |
| "loss": 7.1468, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 5.877413937867338, | |
| "grad_norm": 5.005863666534424, | |
| "learning_rate": 9.192734611503532e-05, | |
| "loss": 7.1239, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.087321578505458, | |
| "grad_norm": 5.807664394378662, | |
| "learning_rate": 9.091826437941474e-05, | |
| "loss": 7.0678, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 6.297229219143577, | |
| "grad_norm": 7.163432598114014, | |
| "learning_rate": 8.990918264379415e-05, | |
| "loss": 7.0203, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.507136859781696, | |
| "grad_norm": 4.169361591339111, | |
| "learning_rate": 8.890010090817356e-05, | |
| "loss": 7.0209, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 6.717044500419815, | |
| "grad_norm": 6.163388252258301, | |
| "learning_rate": 8.789101917255298e-05, | |
| "loss": 7.0227, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 6.926952141057934, | |
| "grad_norm": 4.993062973022461, | |
| "learning_rate": 8.68819374369324e-05, | |
| "loss": 7.0323, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 7.136859781696054, | |
| "grad_norm": 5.657957553863525, | |
| "learning_rate": 8.58728557013118e-05, | |
| "loss": 6.943, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 7.346767422334173, | |
| "grad_norm": 3.4505577087402344, | |
| "learning_rate": 8.486377396569122e-05, | |
| "loss": 6.9042, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 7.5566750629722925, | |
| "grad_norm": 5.620421886444092, | |
| "learning_rate": 8.385469223007064e-05, | |
| "loss": 6.9224, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 7.766582703610411, | |
| "grad_norm": 6.220210552215576, | |
| "learning_rate": 8.284561049445006e-05, | |
| "loss": 6.9333, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 7.97649034424853, | |
| "grad_norm": 4.852828025817871, | |
| "learning_rate": 8.183652875882947e-05, | |
| "loss": 6.9383, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 8.18639798488665, | |
| "grad_norm": 4.6203837394714355, | |
| "learning_rate": 8.082744702320889e-05, | |
| "loss": 6.8173, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 8.39630562552477, | |
| "grad_norm": 4.523601055145264, | |
| "learning_rate": 7.98183652875883e-05, | |
| "loss": 6.8231, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.39630562552477, | |
| "eval_accuracy": 0.15205252672633912, | |
| "eval_loss": 7.090852737426758, | |
| "eval_runtime": 0.6773, | |
| "eval_samples_per_second": 707.199, | |
| "eval_steps_per_second": 7.382, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.606213266162888, | |
| "grad_norm": 5.225585460662842, | |
| "learning_rate": 7.880928355196771e-05, | |
| "loss": 6.8399, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 8.816120906801007, | |
| "grad_norm": 5.009521007537842, | |
| "learning_rate": 7.780020181634713e-05, | |
| "loss": 6.8443, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 9.026028547439127, | |
| "grad_norm": 3.6128814220428467, | |
| "learning_rate": 7.679112008072655e-05, | |
| "loss": 6.8408, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 9.235936188077247, | |
| "grad_norm": 4.911495685577393, | |
| "learning_rate": 7.578203834510595e-05, | |
| "loss": 6.7343, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 9.445843828715365, | |
| "grad_norm": 4.033209800720215, | |
| "learning_rate": 7.477295660948537e-05, | |
| "loss": 6.7516, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 9.655751469353484, | |
| "grad_norm": 5.0978875160217285, | |
| "learning_rate": 7.376387487386479e-05, | |
| "loss": 6.7666, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 9.865659109991604, | |
| "grad_norm": 3.241863965988159, | |
| "learning_rate": 7.27547931382442e-05, | |
| "loss": 6.7687, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 10.075566750629722, | |
| "grad_norm": 5.886169910430908, | |
| "learning_rate": 7.174571140262362e-05, | |
| "loss": 6.733, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 10.285474391267842, | |
| "grad_norm": 3.7957448959350586, | |
| "learning_rate": 7.073662966700304e-05, | |
| "loss": 6.6708, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 10.495382031905962, | |
| "grad_norm": 4.136474609375, | |
| "learning_rate": 6.972754793138244e-05, | |
| "loss": 6.6834, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 10.705289672544081, | |
| "grad_norm": 3.8854360580444336, | |
| "learning_rate": 6.871846619576186e-05, | |
| "loss": 6.6994, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 10.9151973131822, | |
| "grad_norm": 4.119456768035889, | |
| "learning_rate": 6.770938446014128e-05, | |
| "loss": 6.705, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 11.125104953820319, | |
| "grad_norm": 3.676788330078125, | |
| "learning_rate": 6.670030272452068e-05, | |
| "loss": 6.6374, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 11.335012594458439, | |
| "grad_norm": 4.45515775680542, | |
| "learning_rate": 6.56912209889001e-05, | |
| "loss": 6.614, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 11.544920235096557, | |
| "grad_norm": 5.550235271453857, | |
| "learning_rate": 6.468213925327952e-05, | |
| "loss": 6.6212, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 11.754827875734676, | |
| "grad_norm": 5.452288627624512, | |
| "learning_rate": 6.367305751765893e-05, | |
| "loss": 6.6352, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 11.964735516372796, | |
| "grad_norm": 4.133624076843262, | |
| "learning_rate": 6.266397578203835e-05, | |
| "loss": 6.6442, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 12.174643157010916, | |
| "grad_norm": 6.22617244720459, | |
| "learning_rate": 6.165489404641777e-05, | |
| "loss": 6.5665, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 12.384550797649034, | |
| "grad_norm": 4.706435203552246, | |
| "learning_rate": 6.064581231079718e-05, | |
| "loss": 6.5566, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 12.594458438287154, | |
| "grad_norm": 6.133469581604004, | |
| "learning_rate": 5.963673057517659e-05, | |
| "loss": 6.5728, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.594458438287154, | |
| "eval_accuracy": 0.15300668774659318, | |
| "eval_loss": 7.0916547775268555, | |
| "eval_runtime": 0.6727, | |
| "eval_samples_per_second": 712.025, | |
| "eval_steps_per_second": 7.432, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.804366078925273, | |
| "grad_norm": 5.006119728088379, | |
| "learning_rate": 5.8627648839556e-05, | |
| "loss": 6.5821, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 13.014273719563391, | |
| "grad_norm": 4.887792587280273, | |
| "learning_rate": 5.7618567103935416e-05, | |
| "loss": 6.5796, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 13.224181360201511, | |
| "grad_norm": 5.344376564025879, | |
| "learning_rate": 5.660948536831484e-05, | |
| "loss": 6.5008, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 13.43408900083963, | |
| "grad_norm": 3.877203941345215, | |
| "learning_rate": 5.5600403632694254e-05, | |
| "loss": 6.5141, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 13.64399664147775, | |
| "grad_norm": 4.218553066253662, | |
| "learning_rate": 5.4591321897073666e-05, | |
| "loss": 6.5234, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 13.853904282115868, | |
| "grad_norm": 4.164068222045898, | |
| "learning_rate": 5.358224016145308e-05, | |
| "loss": 6.5294, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 14.063811922753988, | |
| "grad_norm": 4.421447277069092, | |
| "learning_rate": 5.257315842583249e-05, | |
| "loss": 6.4964, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 14.273719563392108, | |
| "grad_norm": 4.369233131408691, | |
| "learning_rate": 5.15640766902119e-05, | |
| "loss": 6.4513, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 14.483627204030226, | |
| "grad_norm": 4.740302562713623, | |
| "learning_rate": 5.055499495459133e-05, | |
| "loss": 6.4591, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 14.693534844668346, | |
| "grad_norm": 5.412849426269531, | |
| "learning_rate": 4.954591321897074e-05, | |
| "loss": 6.4735, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 14.903442485306465, | |
| "grad_norm": 4.577197074890137, | |
| "learning_rate": 4.853683148335015e-05, | |
| "loss": 6.4836, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 15.113350125944585, | |
| "grad_norm": 3.7926676273345947, | |
| "learning_rate": 4.7527749747729566e-05, | |
| "loss": 6.4446, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 15.323257766582703, | |
| "grad_norm": 3.7731170654296875, | |
| "learning_rate": 4.6518668012108985e-05, | |
| "loss": 6.4181, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 15.533165407220823, | |
| "grad_norm": 5.084759712219238, | |
| "learning_rate": 4.55095862764884e-05, | |
| "loss": 6.4235, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 15.743073047858942, | |
| "grad_norm": 4.378145694732666, | |
| "learning_rate": 4.450050454086781e-05, | |
| "loss": 6.4324, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 15.95298068849706, | |
| "grad_norm": 6.829603672027588, | |
| "learning_rate": 4.349142280524723e-05, | |
| "loss": 6.427, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 16.162888329135182, | |
| "grad_norm": 3.674001932144165, | |
| "learning_rate": 4.248234106962664e-05, | |
| "loss": 6.3875, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 16.3727959697733, | |
| "grad_norm": 5.228041172027588, | |
| "learning_rate": 4.147325933400605e-05, | |
| "loss": 6.3743, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 16.582703610411418, | |
| "grad_norm": 5.4357781410217285, | |
| "learning_rate": 4.046417759838547e-05, | |
| "loss": 6.3899, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 16.79261125104954, | |
| "grad_norm": 4.4107747077941895, | |
| "learning_rate": 3.9455095862764884e-05, | |
| "loss": 6.3834, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 16.79261125104954, | |
| "eval_accuracy": 0.15201300885948188, | |
| "eval_loss": 7.212457180023193, | |
| "eval_runtime": 0.6729, | |
| "eval_samples_per_second": 711.892, | |
| "eval_steps_per_second": 7.431, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 17.002518891687657, | |
| "grad_norm": 5.2473273277282715, | |
| "learning_rate": 3.84460141271443e-05, | |
| "loss": 6.3931, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 17.212426532325775, | |
| "grad_norm": 6.317814350128174, | |
| "learning_rate": 3.7436932391523715e-05, | |
| "loss": 6.3355, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 17.422334172963897, | |
| "grad_norm": 3.7163403034210205, | |
| "learning_rate": 3.642785065590313e-05, | |
| "loss": 6.3375, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 17.632241813602015, | |
| "grad_norm": 4.958653450012207, | |
| "learning_rate": 3.541876892028255e-05, | |
| "loss": 6.3469, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 17.842149454240133, | |
| "grad_norm": 4.263184070587158, | |
| "learning_rate": 3.440968718466196e-05, | |
| "loss": 6.3539, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 18.052057094878254, | |
| "grad_norm": 4.632514953613281, | |
| "learning_rate": 3.340060544904137e-05, | |
| "loss": 6.3376, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 18.261964735516372, | |
| "grad_norm": 3.5645172595977783, | |
| "learning_rate": 3.239152371342079e-05, | |
| "loss": 6.3026, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 18.471872376154494, | |
| "grad_norm": 5.224117279052734, | |
| "learning_rate": 3.13824419778002e-05, | |
| "loss": 6.3134, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 18.68178001679261, | |
| "grad_norm": 4.207254409790039, | |
| "learning_rate": 3.0373360242179615e-05, | |
| "loss": 6.3148, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 18.89168765743073, | |
| "grad_norm": 3.2214038372039795, | |
| "learning_rate": 2.9364278506559034e-05, | |
| "loss": 6.3143, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 19.10159529806885, | |
| "grad_norm": 3.3819849491119385, | |
| "learning_rate": 2.8355196770938446e-05, | |
| "loss": 6.2998, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 19.31150293870697, | |
| "grad_norm": 4.022982597351074, | |
| "learning_rate": 2.734611503531786e-05, | |
| "loss": 6.2758, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 19.521410579345087, | |
| "grad_norm": 5.50258207321167, | |
| "learning_rate": 2.6337033299697278e-05, | |
| "loss": 6.2822, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 19.73131821998321, | |
| "grad_norm": 3.3523995876312256, | |
| "learning_rate": 2.532795156407669e-05, | |
| "loss": 6.2861, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 19.941225860621326, | |
| "grad_norm": 5.194580078125, | |
| "learning_rate": 2.4318869828456106e-05, | |
| "loss": 6.2814, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 20.151133501259444, | |
| "grad_norm": 3.514719009399414, | |
| "learning_rate": 2.330978809283552e-05, | |
| "loss": 6.2652, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 20.361041141897566, | |
| "grad_norm": 4.825462818145752, | |
| "learning_rate": 2.2300706357214934e-05, | |
| "loss": 6.254, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 20.570948782535684, | |
| "grad_norm": 3.105278968811035, | |
| "learning_rate": 2.129162462159435e-05, | |
| "loss": 6.2578, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 20.7808564231738, | |
| "grad_norm": 3.9636168479919434, | |
| "learning_rate": 2.0282542885973765e-05, | |
| "loss": 6.2572, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 20.990764063811923, | |
| "grad_norm": 4.260512351989746, | |
| "learning_rate": 1.927346115035318e-05, | |
| "loss": 6.2581, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 20.990764063811923, | |
| "eval_accuracy": 0.1502789693554266, | |
| "eval_loss": 7.290200233459473, | |
| "eval_runtime": 0.6732, | |
| "eval_samples_per_second": 711.565, | |
| "eval_steps_per_second": 7.428, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 21.20067170445004, | |
| "grad_norm": 3.9341166019439697, | |
| "learning_rate": 1.8264379414732593e-05, | |
| "loss": 6.2319, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 21.410579345088163, | |
| "grad_norm": 5.869973659515381, | |
| "learning_rate": 1.725529767911201e-05, | |
| "loss": 6.2353, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 21.62048698572628, | |
| "grad_norm": 3.626514196395874, | |
| "learning_rate": 1.6246215943491424e-05, | |
| "loss": 6.2369, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 21.8303946263644, | |
| "grad_norm": 5.164318084716797, | |
| "learning_rate": 1.5237134207870837e-05, | |
| "loss": 6.2347, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 22.04030226700252, | |
| "grad_norm": 3.049724578857422, | |
| "learning_rate": 1.4228052472250252e-05, | |
| "loss": 6.2302, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 22.250209907640638, | |
| "grad_norm": 3.4710233211517334, | |
| "learning_rate": 1.3218970736629668e-05, | |
| "loss": 6.2094, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 22.460117548278756, | |
| "grad_norm": 4.1562018394470215, | |
| "learning_rate": 1.2209889001009082e-05, | |
| "loss": 6.2112, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 22.670025188916878, | |
| "grad_norm": 4.1082282066345215, | |
| "learning_rate": 1.1200807265388496e-05, | |
| "loss": 6.2179, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 22.879932829554996, | |
| "grad_norm": 3.1528453826904297, | |
| "learning_rate": 1.0191725529767912e-05, | |
| "loss": 6.2094, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 23.089840470193113, | |
| "grad_norm": 3.394303321838379, | |
| "learning_rate": 9.182643794147326e-06, | |
| "loss": 6.2048, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 23.299748110831235, | |
| "grad_norm": 3.516995668411255, | |
| "learning_rate": 8.173562058526741e-06, | |
| "loss": 6.1996, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 23.509655751469353, | |
| "grad_norm": 3.9392542839050293, | |
| "learning_rate": 7.164480322906156e-06, | |
| "loss": 6.1992, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 23.719563392107474, | |
| "grad_norm": 3.73042631149292, | |
| "learning_rate": 6.155398587285571e-06, | |
| "loss": 6.1939, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 23.929471032745592, | |
| "grad_norm": 4.406044960021973, | |
| "learning_rate": 5.146316851664986e-06, | |
| "loss": 6.1973, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 24.13937867338371, | |
| "grad_norm": 2.8055338859558105, | |
| "learning_rate": 4.1372351160444e-06, | |
| "loss": 6.1851, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 24.349286314021832, | |
| "grad_norm": 3.4643161296844482, | |
| "learning_rate": 3.1281533804238145e-06, | |
| "loss": 6.189, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 24.55919395465995, | |
| "grad_norm": 4.828442096710205, | |
| "learning_rate": 2.1190716448032293e-06, | |
| "loss": 6.184, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 24.769101595298068, | |
| "grad_norm": 2.78041672706604, | |
| "learning_rate": 1.109989909182644e-06, | |
| "loss": 6.1827, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 24.97900923593619, | |
| "grad_norm": 3.461228132247925, | |
| "learning_rate": 1.0090817356205853e-07, | |
| "loss": 6.1831, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "step": 59550, | |
| "total_flos": 3.058354515764736e+17, | |
| "train_loss": 6.935577958857283, | |
| "train_runtime": 11531.4215, | |
| "train_samples_per_second": 495.687, | |
| "train_steps_per_second": 5.164 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 59550, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 25, | |
| "save_steps": 10000, | |
| "total_flos": 3.058354515764736e+17, | |
| "train_batch_size": 96, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |