| { | |
| "best_metric": 8.311293601989746, | |
| "best_model_checkpoint": "bert_base_lda_20_v1/checkpoint-50000", | |
| "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": 2.09550142288208, | |
| "learning_rate": 5e-06, | |
| "loss": 12.1128, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41981528127623846, | |
| "grad_norm": 1.47028648853302, | |
| "learning_rate": 1e-05, | |
| "loss": 10.3839, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6297229219143576, | |
| "grad_norm": 0.9643331170082092, | |
| "learning_rate": 1.5e-05, | |
| "loss": 9.8774, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8396305625524769, | |
| "grad_norm": 0.9907741546630859, | |
| "learning_rate": 2e-05, | |
| "loss": 9.7389, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0495382031905962, | |
| "grad_norm": 1.2076187133789062, | |
| "learning_rate": 2.5e-05, | |
| "loss": 9.6305, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.2594458438287153, | |
| "grad_norm": 1.0168932676315308, | |
| "learning_rate": 3e-05, | |
| "loss": 9.5318, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4693534844668346, | |
| "grad_norm": 1.077588677406311, | |
| "learning_rate": 3.5e-05, | |
| "loss": 9.4564, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.6792611251049538, | |
| "grad_norm": 1.0115774869918823, | |
| "learning_rate": 4e-05, | |
| "loss": 9.3964, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.8891687657430731, | |
| "grad_norm": 0.9964166283607483, | |
| "learning_rate": 4.5e-05, | |
| "loss": 9.3438, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0990764063811924, | |
| "grad_norm": 1.181398630142212, | |
| "learning_rate": 5e-05, | |
| "loss": 9.296, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.3089840470193117, | |
| "grad_norm": 1.1168142557144165, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 9.2554, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.5188916876574305, | |
| "grad_norm": 1.0137851238250732, | |
| "learning_rate": 6e-05, | |
| "loss": 9.2206, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.72879932829555, | |
| "grad_norm": 1.1866172552108765, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 9.1887, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.938706968933669, | |
| "grad_norm": 0.9899904727935791, | |
| "learning_rate": 7e-05, | |
| "loss": 9.1597, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.1486146095717884, | |
| "grad_norm": 1.0700551271438599, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 9.1129, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.3585222502099077, | |
| "grad_norm": 1.2574646472930908, | |
| "learning_rate": 8e-05, | |
| "loss": 9.0258, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.568429890848027, | |
| "grad_norm": 1.3791582584381104, | |
| "learning_rate": 8.5e-05, | |
| "loss": 8.8912, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.7783375314861463, | |
| "grad_norm": 1.6568559408187866, | |
| "learning_rate": 9e-05, | |
| "loss": 8.7658, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.988245172124265, | |
| "grad_norm": 1.232332468032837, | |
| "learning_rate": 9.5e-05, | |
| "loss": 8.7067, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.198152812762385, | |
| "grad_norm": 1.1055606603622437, | |
| "learning_rate": 0.0001, | |
| "loss": 8.6597, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.198152812762385, | |
| "eval_accuracy": 0.15147889479395643, | |
| "eval_loss": 8.600459098815918, | |
| "eval_runtime": 0.7275, | |
| "eval_samples_per_second": 658.386, | |
| "eval_steps_per_second": 6.873, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.408060453400504, | |
| "grad_norm": 1.2943756580352783, | |
| "learning_rate": 9.899091826437942e-05, | |
| "loss": 8.6387, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.617968094038623, | |
| "grad_norm": 1.1277927160263062, | |
| "learning_rate": 9.798183652875883e-05, | |
| "loss": 8.6152, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.827875734676742, | |
| "grad_norm": 1.249430537223816, | |
| "learning_rate": 9.697275479313825e-05, | |
| "loss": 8.6023, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.037783375314861, | |
| "grad_norm": 1.0872271060943604, | |
| "learning_rate": 9.596367305751767e-05, | |
| "loss": 8.5788, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.247691015952981, | |
| "grad_norm": 1.0338563919067383, | |
| "learning_rate": 9.495459132189707e-05, | |
| "loss": 8.5502, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.4575986565911, | |
| "grad_norm": 1.3109467029571533, | |
| "learning_rate": 9.394550958627649e-05, | |
| "loss": 8.5378, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.667506297229219, | |
| "grad_norm": 1.139695405960083, | |
| "learning_rate": 9.293642785065591e-05, | |
| "loss": 8.5318, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 5.877413937867338, | |
| "grad_norm": 1.239567518234253, | |
| "learning_rate": 9.192734611503532e-05, | |
| "loss": 8.5157, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.087321578505458, | |
| "grad_norm": 1.0034639835357666, | |
| "learning_rate": 9.091826437941474e-05, | |
| "loss": 8.4983, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 6.297229219143577, | |
| "grad_norm": 1.0384876728057861, | |
| "learning_rate": 8.990918264379415e-05, | |
| "loss": 8.4855, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.507136859781696, | |
| "grad_norm": 1.0204209089279175, | |
| "learning_rate": 8.890010090817356e-05, | |
| "loss": 8.4747, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 6.717044500419815, | |
| "grad_norm": 1.1249667406082153, | |
| "learning_rate": 8.789101917255298e-05, | |
| "loss": 8.4679, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 6.926952141057934, | |
| "grad_norm": 0.9850140810012817, | |
| "learning_rate": 8.68819374369324e-05, | |
| "loss": 8.4578, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 7.136859781696054, | |
| "grad_norm": 0.940535843372345, | |
| "learning_rate": 8.58728557013118e-05, | |
| "loss": 8.4404, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 7.346767422334173, | |
| "grad_norm": 0.9573142528533936, | |
| "learning_rate": 8.486377396569122e-05, | |
| "loss": 8.4296, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 7.5566750629722925, | |
| "grad_norm": 1.068617582321167, | |
| "learning_rate": 8.385469223007064e-05, | |
| "loss": 8.4299, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 7.766582703610411, | |
| "grad_norm": 0.9672011733055115, | |
| "learning_rate": 8.284561049445006e-05, | |
| "loss": 8.4241, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 7.97649034424853, | |
| "grad_norm": 1.0434781312942505, | |
| "learning_rate": 8.183652875882947e-05, | |
| "loss": 8.4203, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 8.18639798488665, | |
| "grad_norm": 0.9273554682731628, | |
| "learning_rate": 8.082744702320889e-05, | |
| "loss": 8.3997, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 8.39630562552477, | |
| "grad_norm": 1.0744283199310303, | |
| "learning_rate": 7.98183652875883e-05, | |
| "loss": 8.3979, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.39630562552477, | |
| "eval_accuracy": 0.1533993658632397, | |
| "eval_loss": 8.40470027923584, | |
| "eval_runtime": 0.7704, | |
| "eval_samples_per_second": 621.764, | |
| "eval_steps_per_second": 6.49, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.606213266162888, | |
| "grad_norm": 1.064097285270691, | |
| "learning_rate": 7.880928355196771e-05, | |
| "loss": 8.3931, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 8.816120906801007, | |
| "grad_norm": 0.9334086179733276, | |
| "learning_rate": 7.780020181634713e-05, | |
| "loss": 8.391, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 9.026028547439127, | |
| "grad_norm": 0.994073212146759, | |
| "learning_rate": 7.679112008072655e-05, | |
| "loss": 8.3826, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 9.235936188077247, | |
| "grad_norm": 1.0116281509399414, | |
| "learning_rate": 7.578203834510595e-05, | |
| "loss": 8.3702, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 9.445843828715365, | |
| "grad_norm": 0.9388275146484375, | |
| "learning_rate": 7.477295660948537e-05, | |
| "loss": 8.3661, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 9.655751469353484, | |
| "grad_norm": 0.906207799911499, | |
| "learning_rate": 7.376387487386479e-05, | |
| "loss": 8.3642, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 9.865659109991604, | |
| "grad_norm": 0.9428483843803406, | |
| "learning_rate": 7.27547931382442e-05, | |
| "loss": 8.3596, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 10.075566750629722, | |
| "grad_norm": 0.9630805850028992, | |
| "learning_rate": 7.174571140262362e-05, | |
| "loss": 8.3472, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 10.285474391267842, | |
| "grad_norm": 0.8972500562667847, | |
| "learning_rate": 7.073662966700304e-05, | |
| "loss": 8.3477, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 10.495382031905962, | |
| "grad_norm": 0.9225510358810425, | |
| "learning_rate": 6.972754793138244e-05, | |
| "loss": 8.3384, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 10.705289672544081, | |
| "grad_norm": 0.9384250044822693, | |
| "learning_rate": 6.871846619576186e-05, | |
| "loss": 8.342, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 10.9151973131822, | |
| "grad_norm": 0.9212971925735474, | |
| "learning_rate": 6.770938446014128e-05, | |
| "loss": 8.3399, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 11.125104953820319, | |
| "grad_norm": 0.9395396709442139, | |
| "learning_rate": 6.670030272452068e-05, | |
| "loss": 8.33, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 11.335012594458439, | |
| "grad_norm": 0.8887066841125488, | |
| "learning_rate": 6.56912209889001e-05, | |
| "loss": 8.3248, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 11.544920235096557, | |
| "grad_norm": 0.8518480062484741, | |
| "learning_rate": 6.468213925327952e-05, | |
| "loss": 8.3184, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 11.754827875734676, | |
| "grad_norm": 0.9201586246490479, | |
| "learning_rate": 6.367305751765893e-05, | |
| "loss": 8.3152, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 11.964735516372796, | |
| "grad_norm": 0.9495044946670532, | |
| "learning_rate": 6.266397578203835e-05, | |
| "loss": 8.3148, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 12.174643157010916, | |
| "grad_norm": 0.8486906886100769, | |
| "learning_rate": 6.165489404641777e-05, | |
| "loss": 8.308, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 12.384550797649034, | |
| "grad_norm": 0.9100642204284668, | |
| "learning_rate": 6.064581231079718e-05, | |
| "loss": 8.2981, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 12.594458438287154, | |
| "grad_norm": 0.8565881252288818, | |
| "learning_rate": 5.963673057517659e-05, | |
| "loss": 8.3034, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.594458438287154, | |
| "eval_accuracy": 0.15485351316562665, | |
| "eval_loss": 8.349508285522461, | |
| "eval_runtime": 0.7093, | |
| "eval_samples_per_second": 675.332, | |
| "eval_steps_per_second": 7.049, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.804366078925273, | |
| "grad_norm": 0.8767760992050171, | |
| "learning_rate": 5.8627648839556e-05, | |
| "loss": 8.2988, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 13.014273719563391, | |
| "grad_norm": 0.8581773042678833, | |
| "learning_rate": 5.7618567103935416e-05, | |
| "loss": 8.3012, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 13.224181360201511, | |
| "grad_norm": 0.8876222968101501, | |
| "learning_rate": 5.660948536831484e-05, | |
| "loss": 8.2868, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 13.43408900083963, | |
| "grad_norm": 0.8816433548927307, | |
| "learning_rate": 5.5600403632694254e-05, | |
| "loss": 8.2865, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 13.64399664147775, | |
| "grad_norm": 0.8398271203041077, | |
| "learning_rate": 5.4591321897073666e-05, | |
| "loss": 8.2892, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 13.853904282115868, | |
| "grad_norm": 0.9173211455345154, | |
| "learning_rate": 5.358224016145308e-05, | |
| "loss": 8.2846, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 14.063811922753988, | |
| "grad_norm": 0.906325101852417, | |
| "learning_rate": 5.257315842583249e-05, | |
| "loss": 8.2806, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 14.273719563392108, | |
| "grad_norm": 0.8287147283554077, | |
| "learning_rate": 5.15640766902119e-05, | |
| "loss": 8.2719, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 14.483627204030226, | |
| "grad_norm": 0.8609397411346436, | |
| "learning_rate": 5.055499495459133e-05, | |
| "loss": 8.2716, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 14.693534844668346, | |
| "grad_norm": 0.9240869879722595, | |
| "learning_rate": 4.954591321897074e-05, | |
| "loss": 8.2722, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 14.903442485306465, | |
| "grad_norm": 0.8659062385559082, | |
| "learning_rate": 4.853683148335015e-05, | |
| "loss": 8.2708, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 15.113350125944585, | |
| "grad_norm": 0.8485491275787354, | |
| "learning_rate": 4.7527749747729566e-05, | |
| "loss": 8.2624, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 15.323257766582703, | |
| "grad_norm": 0.8944738507270813, | |
| "learning_rate": 4.6518668012108985e-05, | |
| "loss": 8.2607, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 15.533165407220823, | |
| "grad_norm": 0.8862741589546204, | |
| "learning_rate": 4.55095862764884e-05, | |
| "loss": 8.2568, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 15.743073047858942, | |
| "grad_norm": 0.8353709578514099, | |
| "learning_rate": 4.450050454086781e-05, | |
| "loss": 8.2601, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 15.95298068849706, | |
| "grad_norm": 0.8464979529380798, | |
| "learning_rate": 4.349142280524723e-05, | |
| "loss": 8.2562, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 16.162888329135182, | |
| "grad_norm": 0.8389509320259094, | |
| "learning_rate": 4.248234106962664e-05, | |
| "loss": 8.2529, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 16.3727959697733, | |
| "grad_norm": 0.8426734805107117, | |
| "learning_rate": 4.147325933400605e-05, | |
| "loss": 8.2448, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 16.582703610411418, | |
| "grad_norm": 0.8700066804885864, | |
| "learning_rate": 4.046417759838547e-05, | |
| "loss": 8.2489, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 16.79261125104954, | |
| "grad_norm": 0.9079403281211853, | |
| "learning_rate": 3.9455095862764884e-05, | |
| "loss": 8.239, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 16.79261125104954, | |
| "eval_accuracy": 0.15366715262980823, | |
| "eval_loss": 8.336407661437988, | |
| "eval_runtime": 0.7881, | |
| "eval_samples_per_second": 607.798, | |
| "eval_steps_per_second": 6.344, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 17.002518891687657, | |
| "grad_norm": 0.875669002532959, | |
| "learning_rate": 3.84460141271443e-05, | |
| "loss": 8.2462, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 17.212426532325775, | |
| "grad_norm": 0.8467079401016235, | |
| "learning_rate": 3.7436932391523715e-05, | |
| "loss": 8.2352, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 17.422334172963897, | |
| "grad_norm": 0.9035972952842712, | |
| "learning_rate": 3.642785065590313e-05, | |
| "loss": 8.2328, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 17.632241813602015, | |
| "grad_norm": 0.8857328295707703, | |
| "learning_rate": 3.541876892028255e-05, | |
| "loss": 8.236, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 17.842149454240133, | |
| "grad_norm": 0.8668442964553833, | |
| "learning_rate": 3.440968718466196e-05, | |
| "loss": 8.2352, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 18.052057094878254, | |
| "grad_norm": 0.9012500047683716, | |
| "learning_rate": 3.340060544904137e-05, | |
| "loss": 8.2293, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 18.261964735516372, | |
| "grad_norm": 0.8686524033546448, | |
| "learning_rate": 3.239152371342079e-05, | |
| "loss": 8.2237, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 18.471872376154494, | |
| "grad_norm": 0.8950616717338562, | |
| "learning_rate": 3.13824419778002e-05, | |
| "loss": 8.2235, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 18.68178001679261, | |
| "grad_norm": 0.8709149360656738, | |
| "learning_rate": 3.0373360242179615e-05, | |
| "loss": 8.2252, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 18.89168765743073, | |
| "grad_norm": 0.8441091179847717, | |
| "learning_rate": 2.9364278506559034e-05, | |
| "loss": 8.2264, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 19.10159529806885, | |
| "grad_norm": 0.8468266725540161, | |
| "learning_rate": 2.8355196770938446e-05, | |
| "loss": 8.2224, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 19.31150293870697, | |
| "grad_norm": 0.8458043932914734, | |
| "learning_rate": 2.734611503531786e-05, | |
| "loss": 8.2098, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 19.521410579345087, | |
| "grad_norm": 0.8512638807296753, | |
| "learning_rate": 2.6337033299697278e-05, | |
| "loss": 8.2156, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 19.73131821998321, | |
| "grad_norm": 0.8596357703208923, | |
| "learning_rate": 2.532795156407669e-05, | |
| "loss": 8.2162, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 19.941225860621326, | |
| "grad_norm": 0.863677978515625, | |
| "learning_rate": 2.4318869828456106e-05, | |
| "loss": 8.2119, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 20.151133501259444, | |
| "grad_norm": 0.8351937532424927, | |
| "learning_rate": 2.330978809283552e-05, | |
| "loss": 8.2071, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 20.361041141897566, | |
| "grad_norm": 0.8820331692695618, | |
| "learning_rate": 2.2300706357214934e-05, | |
| "loss": 8.2066, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 20.570948782535684, | |
| "grad_norm": 0.8816234469413757, | |
| "learning_rate": 2.129162462159435e-05, | |
| "loss": 8.2035, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 20.7808564231738, | |
| "grad_norm": 0.8651553392410278, | |
| "learning_rate": 2.0282542885973765e-05, | |
| "loss": 8.203, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 20.990764063811923, | |
| "grad_norm": 0.8483617305755615, | |
| "learning_rate": 1.927346115035318e-05, | |
| "loss": 8.2045, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 20.990764063811923, | |
| "eval_accuracy": 0.15238175344155663, | |
| "eval_loss": 8.311293601989746, | |
| "eval_runtime": 0.701, | |
| "eval_samples_per_second": 683.323, | |
| "eval_steps_per_second": 7.133, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 21.20067170445004, | |
| "grad_norm": 0.8452841639518738, | |
| "learning_rate": 1.8264379414732593e-05, | |
| "loss": 8.1973, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 21.410579345088163, | |
| "grad_norm": 0.8958883285522461, | |
| "learning_rate": 1.725529767911201e-05, | |
| "loss": 8.1981, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 21.62048698572628, | |
| "grad_norm": 0.899661123752594, | |
| "learning_rate": 1.6246215943491424e-05, | |
| "loss": 8.1963, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 21.8303946263644, | |
| "grad_norm": 0.8597267866134644, | |
| "learning_rate": 1.5237134207870837e-05, | |
| "loss": 8.1965, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 22.04030226700252, | |
| "grad_norm": 0.8393648862838745, | |
| "learning_rate": 1.4228052472250252e-05, | |
| "loss": 8.1953, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 22.250209907640638, | |
| "grad_norm": 0.8742464780807495, | |
| "learning_rate": 1.3218970736629668e-05, | |
| "loss": 8.1841, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 22.460117548278756, | |
| "grad_norm": 0.8778883814811707, | |
| "learning_rate": 1.2209889001009082e-05, | |
| "loss": 8.1898, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 22.670025188916878, | |
| "grad_norm": 0.8447439074516296, | |
| "learning_rate": 1.1200807265388496e-05, | |
| "loss": 8.1904, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 22.879932829554996, | |
| "grad_norm": 0.8448654413223267, | |
| "learning_rate": 1.0191725529767912e-05, | |
| "loss": 8.1873, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 23.089840470193113, | |
| "grad_norm": 0.8719317317008972, | |
| "learning_rate": 9.182643794147326e-06, | |
| "loss": 8.1851, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 23.299748110831235, | |
| "grad_norm": 0.8342525362968445, | |
| "learning_rate": 8.173562058526741e-06, | |
| "loss": 8.1828, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 23.509655751469353, | |
| "grad_norm": 0.9656890034675598, | |
| "learning_rate": 7.164480322906156e-06, | |
| "loss": 8.1706, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 23.719563392107474, | |
| "grad_norm": 0.9281942248344421, | |
| "learning_rate": 6.155398587285571e-06, | |
| "loss": 8.1148, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 23.929471032745592, | |
| "grad_norm": 0.9900497198104858, | |
| "learning_rate": 5.146316851664986e-06, | |
| "loss": 8.09, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 24.13937867338371, | |
| "grad_norm": 1.0370519161224365, | |
| "learning_rate": 4.1372351160444e-06, | |
| "loss": 8.0687, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 24.349286314021832, | |
| "grad_norm": 1.0472075939178467, | |
| "learning_rate": 3.1281533804238145e-06, | |
| "loss": 8.0603, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 24.55919395465995, | |
| "grad_norm": 1.0631659030914307, | |
| "learning_rate": 2.1190716448032293e-06, | |
| "loss": 8.0468, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 24.769101595298068, | |
| "grad_norm": 0.9905564785003662, | |
| "learning_rate": 1.109989909182644e-06, | |
| "loss": 8.0438, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 24.97900923593619, | |
| "grad_norm": 1.0178414583206177, | |
| "learning_rate": 1.0090817356205853e-07, | |
| "loss": 8.0397, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "step": 59550, | |
| "total_flos": 1.5151126801503744e+18, | |
| "train_loss": 8.487465922806464, | |
| "train_runtime": 17081.4633, | |
| "train_samples_per_second": 334.63, | |
| "train_steps_per_second": 3.486 | |
| } | |
| ], | |
| "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": 1.5151126801503744e+18, | |
| "train_batch_size": 96, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |