| { | |
| "best_metric": 7.017268180847168, | |
| "best_model_checkpoint": "bert_base_lda_100_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": 2.099632501602173, | |
| "learning_rate": 5e-06, | |
| "loss": 13.7106, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41981528127623846, | |
| "grad_norm": 1.3332628011703491, | |
| "learning_rate": 1e-05, | |
| "loss": 11.963, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6297229219143576, | |
| "grad_norm": 0.9692000150680542, | |
| "learning_rate": 1.5e-05, | |
| "loss": 11.4559, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8396305625524769, | |
| "grad_norm": 0.9592443704605103, | |
| "learning_rate": 2e-05, | |
| "loss": 11.3196, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0495382031905962, | |
| "grad_norm": 1.16844642162323, | |
| "learning_rate": 2.5e-05, | |
| "loss": 11.2143, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.2594458438287153, | |
| "grad_norm": 0.9776472449302673, | |
| "learning_rate": 3e-05, | |
| "loss": 11.1187, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4693534844668346, | |
| "grad_norm": 1.1129740476608276, | |
| "learning_rate": 3.5e-05, | |
| "loss": 11.0441, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.6792611251049538, | |
| "grad_norm": 0.8808595538139343, | |
| "learning_rate": 4e-05, | |
| "loss": 10.9835, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.8891687657430731, | |
| "grad_norm": 0.930133581161499, | |
| "learning_rate": 4.5e-05, | |
| "loss": 10.9304, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0990764063811924, | |
| "grad_norm": 0.9473118782043457, | |
| "learning_rate": 5e-05, | |
| "loss": 10.8812, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.3089840470193117, | |
| "grad_norm": 1.0284568071365356, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 10.8399, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.5188916876574305, | |
| "grad_norm": 0.898517906665802, | |
| "learning_rate": 6e-05, | |
| "loss": 10.8057, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.72879932829555, | |
| "grad_norm": 0.9422119855880737, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 10.7751, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.938706968933669, | |
| "grad_norm": 0.8704677224159241, | |
| "learning_rate": 7e-05, | |
| "loss": 10.7466, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.1486146095717884, | |
| "grad_norm": 0.8451219201087952, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 10.7197, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.3585222502099077, | |
| "grad_norm": 0.8918795585632324, | |
| "learning_rate": 8e-05, | |
| "loss": 10.6978, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.568429890848027, | |
| "grad_norm": 0.9373081922531128, | |
| "learning_rate": 8.5e-05, | |
| "loss": 10.6818, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.7783375314861463, | |
| "grad_norm": 0.8285127282142639, | |
| "learning_rate": 9e-05, | |
| "loss": 10.6535, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.988245172124265, | |
| "grad_norm": 0.9265228509902954, | |
| "learning_rate": 9.5e-05, | |
| "loss": 10.6104, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.198152812762385, | |
| "grad_norm": 0.9272535443305969, | |
| "learning_rate": 0.0001, | |
| "loss": 10.4589, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.198152812762385, | |
| "eval_accuracy": 0.1509780461337266, | |
| "eval_loss": 10.293501853942871, | |
| "eval_runtime": 0.7304, | |
| "eval_samples_per_second": 655.812, | |
| "eval_steps_per_second": 6.846, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.408060453400504, | |
| "grad_norm": 1.1058239936828613, | |
| "learning_rate": 9.899091826437942e-05, | |
| "loss": 10.2508, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.617968094038623, | |
| "grad_norm": 1.1870856285095215, | |
| "learning_rate": 9.798183652875883e-05, | |
| "loss": 10.0973, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.827875734676742, | |
| "grad_norm": 1.0980850458145142, | |
| "learning_rate": 9.697275479313825e-05, | |
| "loss": 9.9933, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 5.037783375314861, | |
| "grad_norm": 1.1159865856170654, | |
| "learning_rate": 9.596367305751767e-05, | |
| "loss": 9.9118, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.247691015952981, | |
| "grad_norm": 1.093456745147705, | |
| "learning_rate": 9.495459132189707e-05, | |
| "loss": 9.8527, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.4575986565911, | |
| "grad_norm": 1.0816402435302734, | |
| "learning_rate": 9.394550958627649e-05, | |
| "loss": 9.8213, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.667506297229219, | |
| "grad_norm": 1.061692476272583, | |
| "learning_rate": 9.293642785065591e-05, | |
| "loss": 9.8016, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 5.877413937867338, | |
| "grad_norm": 1.1223222017288208, | |
| "learning_rate": 9.192734611503532e-05, | |
| "loss": 9.7785, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 6.087321578505458, | |
| "grad_norm": 1.0292482376098633, | |
| "learning_rate": 9.091826437941474e-05, | |
| "loss": 9.7528, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 6.297229219143577, | |
| "grad_norm": 1.096860647201538, | |
| "learning_rate": 8.990918264379415e-05, | |
| "loss": 9.7305, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.507136859781696, | |
| "grad_norm": 1.0757558345794678, | |
| "learning_rate": 8.890010090817356e-05, | |
| "loss": 9.7141, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 6.717044500419815, | |
| "grad_norm": 1.0448325872421265, | |
| "learning_rate": 8.789101917255298e-05, | |
| "loss": 9.7031, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 6.926952141057934, | |
| "grad_norm": 1.055416464805603, | |
| "learning_rate": 8.68819374369324e-05, | |
| "loss": 9.6886, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 7.136859781696054, | |
| "grad_norm": 1.0751450061798096, | |
| "learning_rate": 8.58728557013118e-05, | |
| "loss": 9.6648, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 7.346767422334173, | |
| "grad_norm": 1.017815113067627, | |
| "learning_rate": 8.486377396569122e-05, | |
| "loss": 9.6503, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 7.5566750629722925, | |
| "grad_norm": 1.0497677326202393, | |
| "learning_rate": 8.385469223007064e-05, | |
| "loss": 9.6478, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 7.766582703610411, | |
| "grad_norm": 1.082327127456665, | |
| "learning_rate": 8.284561049445006e-05, | |
| "loss": 9.6374, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 7.97649034424853, | |
| "grad_norm": 1.025443196296692, | |
| "learning_rate": 8.183652875882947e-05, | |
| "loss": 9.6337, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 8.18639798488665, | |
| "grad_norm": 1.0154904127120972, | |
| "learning_rate": 8.082744702320889e-05, | |
| "loss": 9.6079, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 8.39630562552477, | |
| "grad_norm": 1.035565733909607, | |
| "learning_rate": 7.98183652875883e-05, | |
| "loss": 9.6043, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.39630562552477, | |
| "eval_accuracy": 0.15247341395662056, | |
| "eval_loss": 9.617895126342773, | |
| "eval_runtime": 0.7058, | |
| "eval_samples_per_second": 678.693, | |
| "eval_steps_per_second": 7.084, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.606213266162888, | |
| "grad_norm": 1.037238597869873, | |
| "learning_rate": 7.880928355196771e-05, | |
| "loss": 9.5952, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 8.816120906801007, | |
| "grad_norm": 0.9944680333137512, | |
| "learning_rate": 7.780020181634713e-05, | |
| "loss": 9.593, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 9.026028547439127, | |
| "grad_norm": 1.041021466255188, | |
| "learning_rate": 7.679112008072655e-05, | |
| "loss": 9.5821, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 9.235936188077247, | |
| "grad_norm": 1.036467432975769, | |
| "learning_rate": 7.578203834510595e-05, | |
| "loss": 9.5663, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 9.445843828715365, | |
| "grad_norm": 1.0199445486068726, | |
| "learning_rate": 7.477295660948537e-05, | |
| "loss": 9.5626, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 9.655751469353484, | |
| "grad_norm": 1.0167254209518433, | |
| "learning_rate": 7.376387487386479e-05, | |
| "loss": 9.5583, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 9.865659109991604, | |
| "grad_norm": 1.0065767765045166, | |
| "learning_rate": 7.27547931382442e-05, | |
| "loss": 9.553, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 10.075566750629722, | |
| "grad_norm": 0.9897018671035767, | |
| "learning_rate": 7.174571140262362e-05, | |
| "loss": 9.538, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 10.285474391267842, | |
| "grad_norm": 0.9982227087020874, | |
| "learning_rate": 7.073662966700304e-05, | |
| "loss": 9.536, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 10.495382031905962, | |
| "grad_norm": 0.9569253921508789, | |
| "learning_rate": 6.972754793138244e-05, | |
| "loss": 9.5271, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 10.705289672544081, | |
| "grad_norm": 0.9898385405540466, | |
| "learning_rate": 6.871846619576186e-05, | |
| "loss": 9.5276, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 10.9151973131822, | |
| "grad_norm": 1.0395140647888184, | |
| "learning_rate": 6.770938446014128e-05, | |
| "loss": 9.5282, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 11.125104953820319, | |
| "grad_norm": 1.00431227684021, | |
| "learning_rate": 6.670030272452068e-05, | |
| "loss": 9.5134, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 11.335012594458439, | |
| "grad_norm": 0.9317150712013245, | |
| "learning_rate": 6.56912209889001e-05, | |
| "loss": 9.5057, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 11.544920235096557, | |
| "grad_norm": 0.9892732501029968, | |
| "learning_rate": 6.468213925327952e-05, | |
| "loss": 9.5005, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 11.754827875734676, | |
| "grad_norm": 1.0242632627487183, | |
| "learning_rate": 6.367305751765893e-05, | |
| "loss": 9.4975, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 11.964735516372796, | |
| "grad_norm": 1.0226876735687256, | |
| "learning_rate": 6.266397578203835e-05, | |
| "loss": 9.4968, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 12.174643157010916, | |
| "grad_norm": 0.984201967716217, | |
| "learning_rate": 6.165489404641777e-05, | |
| "loss": 9.4866, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 12.384550797649034, | |
| "grad_norm": 0.9562113285064697, | |
| "learning_rate": 6.064581231079718e-05, | |
| "loss": 9.4767, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 12.594458438287154, | |
| "grad_norm": 0.9997998476028442, | |
| "learning_rate": 5.963673057517659e-05, | |
| "loss": 9.48, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.594458438287154, | |
| "eval_accuracy": 0.156084730111649, | |
| "eval_loss": 9.544936180114746, | |
| "eval_runtime": 0.7085, | |
| "eval_samples_per_second": 676.096, | |
| "eval_steps_per_second": 7.057, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 12.804366078925273, | |
| "grad_norm": 0.988662600517273, | |
| "learning_rate": 5.8627648839556e-05, | |
| "loss": 9.4762, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 13.014273719563391, | |
| "grad_norm": 0.9693503379821777, | |
| "learning_rate": 5.7618567103935416e-05, | |
| "loss": 9.4768, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 13.224181360201511, | |
| "grad_norm": 0.9838449358940125, | |
| "learning_rate": 5.660948536831484e-05, | |
| "loss": 9.4605, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 13.43408900083963, | |
| "grad_norm": 0.985416054725647, | |
| "learning_rate": 5.5600403632694254e-05, | |
| "loss": 9.4605, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 13.64399664147775, | |
| "grad_norm": 0.9869274497032166, | |
| "learning_rate": 5.4591321897073666e-05, | |
| "loss": 9.4617, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 13.853904282115868, | |
| "grad_norm": 0.9963878393173218, | |
| "learning_rate": 5.358224016145308e-05, | |
| "loss": 9.4569, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 14.063811922753988, | |
| "grad_norm": 0.9579880833625793, | |
| "learning_rate": 5.257315842583249e-05, | |
| "loss": 9.4508, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 14.273719563392108, | |
| "grad_norm": 0.9933662414550781, | |
| "learning_rate": 5.15640766902119e-05, | |
| "loss": 9.4407, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 14.483627204030226, | |
| "grad_norm": 0.966357409954071, | |
| "learning_rate": 5.055499495459133e-05, | |
| "loss": 9.4404, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 14.693534844668346, | |
| "grad_norm": 1.005331039428711, | |
| "learning_rate": 4.954591321897074e-05, | |
| "loss": 9.4389, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 14.903442485306465, | |
| "grad_norm": 0.9809695482254028, | |
| "learning_rate": 4.853683148335015e-05, | |
| "loss": 9.4387, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 15.113350125944585, | |
| "grad_norm": 0.954156219959259, | |
| "learning_rate": 4.7527749747729566e-05, | |
| "loss": 9.4283, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 15.323257766582703, | |
| "grad_norm": 1.000462293624878, | |
| "learning_rate": 4.6518668012108985e-05, | |
| "loss": 9.423, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 15.533165407220823, | |
| "grad_norm": 0.9715126752853394, | |
| "learning_rate": 4.55095862764884e-05, | |
| "loss": 9.4215, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 15.743073047858942, | |
| "grad_norm": 0.9889194369316101, | |
| "learning_rate": 4.450050454086781e-05, | |
| "loss": 9.4211, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 15.95298068849706, | |
| "grad_norm": 0.9900258183479309, | |
| "learning_rate": 4.349142280524723e-05, | |
| "loss": 9.4147, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 16.162888329135182, | |
| "grad_norm": 0.9751452207565308, | |
| "learning_rate": 4.248234106962664e-05, | |
| "loss": 9.3913, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 16.3727959697733, | |
| "grad_norm": 1.0791572332382202, | |
| "learning_rate": 4.147325933400605e-05, | |
| "loss": 9.3417, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 16.582703610411418, | |
| "grad_norm": 1.1733478307724, | |
| "learning_rate": 4.046417759838547e-05, | |
| "loss": 9.2246, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 16.79261125104954, | |
| "grad_norm": 1.2456886768341064, | |
| "learning_rate": 3.9455095862764884e-05, | |
| "loss": 8.9658, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 16.79261125104954, | |
| "eval_accuracy": 0.23029045643153526, | |
| "eval_loss": 8.832155227661133, | |
| "eval_runtime": 0.7062, | |
| "eval_samples_per_second": 678.27, | |
| "eval_steps_per_second": 7.08, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 17.002518891687657, | |
| "grad_norm": 1.308842658996582, | |
| "learning_rate": 3.84460141271443e-05, | |
| "loss": 8.7837, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 17.212426532325775, | |
| "grad_norm": 1.3732590675354004, | |
| "learning_rate": 3.7436932391523715e-05, | |
| "loss": 8.6164, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 17.422334172963897, | |
| "grad_norm": 1.4219366312026978, | |
| "learning_rate": 3.642785065590313e-05, | |
| "loss": 8.4975, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 17.632241813602015, | |
| "grad_norm": 1.481255292892456, | |
| "learning_rate": 3.541876892028255e-05, | |
| "loss": 8.4114, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 17.842149454240133, | |
| "grad_norm": 1.408211588859558, | |
| "learning_rate": 3.440968718466196e-05, | |
| "loss": 8.3358, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 18.052057094878254, | |
| "grad_norm": 1.456390619277954, | |
| "learning_rate": 3.340060544904137e-05, | |
| "loss": 8.2576, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 18.261964735516372, | |
| "grad_norm": 1.5487608909606934, | |
| "learning_rate": 3.239152371342079e-05, | |
| "loss": 8.1819, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 18.471872376154494, | |
| "grad_norm": 1.5488317012786865, | |
| "learning_rate": 3.13824419778002e-05, | |
| "loss": 8.1052, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 18.68178001679261, | |
| "grad_norm": 1.4640696048736572, | |
| "learning_rate": 3.0373360242179615e-05, | |
| "loss": 7.9731, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 18.89168765743073, | |
| "grad_norm": 1.416163682937622, | |
| "learning_rate": 2.9364278506559034e-05, | |
| "loss": 7.8276, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 19.10159529806885, | |
| "grad_norm": 1.3770819902420044, | |
| "learning_rate": 2.8355196770938446e-05, | |
| "loss": 7.7208, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 19.31150293870697, | |
| "grad_norm": 1.4795335531234741, | |
| "learning_rate": 2.734611503531786e-05, | |
| "loss": 7.6359, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 19.521410579345087, | |
| "grad_norm": 1.4281933307647705, | |
| "learning_rate": 2.6337033299697278e-05, | |
| "loss": 7.5747, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 19.73131821998321, | |
| "grad_norm": 1.414133906364441, | |
| "learning_rate": 2.532795156407669e-05, | |
| "loss": 7.5191, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 19.941225860621326, | |
| "grad_norm": 1.489829182624817, | |
| "learning_rate": 2.4318869828456106e-05, | |
| "loss": 7.4639, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 20.151133501259444, | |
| "grad_norm": 1.4401626586914062, | |
| "learning_rate": 2.330978809283552e-05, | |
| "loss": 7.4126, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 20.361041141897566, | |
| "grad_norm": 1.5048668384552002, | |
| "learning_rate": 2.2300706357214934e-05, | |
| "loss": 7.3737, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 20.570948782535684, | |
| "grad_norm": 1.5247819423675537, | |
| "learning_rate": 2.129162462159435e-05, | |
| "loss": 7.3287, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 20.7808564231738, | |
| "grad_norm": 1.531874179840088, | |
| "learning_rate": 2.0282542885973765e-05, | |
| "loss": 7.2904, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 20.990764063811923, | |
| "grad_norm": 1.5145772695541382, | |
| "learning_rate": 1.927346115035318e-05, | |
| "loss": 7.2614, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 20.990764063811923, | |
| "eval_accuracy": 0.4201082232876328, | |
| "eval_loss": 7.017268180847168, | |
| "eval_runtime": 0.7142, | |
| "eval_samples_per_second": 670.716, | |
| "eval_steps_per_second": 7.001, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 21.20067170445004, | |
| "grad_norm": 1.5500428676605225, | |
| "learning_rate": 1.8264379414732593e-05, | |
| "loss": 7.2181, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 21.410579345088163, | |
| "grad_norm": 1.5977363586425781, | |
| "learning_rate": 1.725529767911201e-05, | |
| "loss": 7.1958, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 21.62048698572628, | |
| "grad_norm": 1.5731196403503418, | |
| "learning_rate": 1.6246215943491424e-05, | |
| "loss": 7.1656, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 21.8303946263644, | |
| "grad_norm": 1.6346486806869507, | |
| "learning_rate": 1.5237134207870837e-05, | |
| "loss": 7.1388, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 22.04030226700252, | |
| "grad_norm": 1.6242817640304565, | |
| "learning_rate": 1.4228052472250252e-05, | |
| "loss": 7.1167, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 22.250209907640638, | |
| "grad_norm": 1.6085208654403687, | |
| "learning_rate": 1.3218970736629668e-05, | |
| "loss": 7.0882, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 22.460117548278756, | |
| "grad_norm": 1.5465388298034668, | |
| "learning_rate": 1.2209889001009082e-05, | |
| "loss": 7.0714, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 22.670025188916878, | |
| "grad_norm": 1.604613184928894, | |
| "learning_rate": 1.1200807265388496e-05, | |
| "loss": 7.0556, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 22.879932829554996, | |
| "grad_norm": 1.6013134717941284, | |
| "learning_rate": 1.0191725529767912e-05, | |
| "loss": 7.0381, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 23.089840470193113, | |
| "grad_norm": 1.671255111694336, | |
| "learning_rate": 9.182643794147326e-06, | |
| "loss": 7.0206, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 23.299748110831235, | |
| "grad_norm": 1.6133580207824707, | |
| "learning_rate": 8.173562058526741e-06, | |
| "loss": 7.0051, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 23.509655751469353, | |
| "grad_norm": 1.6531249284744263, | |
| "learning_rate": 7.164480322906156e-06, | |
| "loss": 6.9929, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 23.719563392107474, | |
| "grad_norm": 1.6123356819152832, | |
| "learning_rate": 6.155398587285571e-06, | |
| "loss": 6.9788, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 23.929471032745592, | |
| "grad_norm": 1.5911654233932495, | |
| "learning_rate": 5.146316851664986e-06, | |
| "loss": 6.9751, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 24.13937867338371, | |
| "grad_norm": 1.6447371244430542, | |
| "learning_rate": 4.1372351160444e-06, | |
| "loss": 6.9654, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 24.349286314021832, | |
| "grad_norm": 1.599005103111267, | |
| "learning_rate": 3.1281533804238145e-06, | |
| "loss": 6.9591, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 24.55919395465995, | |
| "grad_norm": 1.6058322191238403, | |
| "learning_rate": 2.1190716448032293e-06, | |
| "loss": 6.9498, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 24.769101595298068, | |
| "grad_norm": 1.5888394117355347, | |
| "learning_rate": 1.109989909182644e-06, | |
| "loss": 6.9459, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 24.97900923593619, | |
| "grad_norm": 1.6043498516082764, | |
| "learning_rate": 1.0090817356205853e-07, | |
| "loss": 6.9472, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "step": 59550, | |
| "total_flos": 1.5161929390646784e+18, | |
| "train_loss": 9.134388991963654, | |
| "train_runtime": 17204.0061, | |
| "train_samples_per_second": 332.247, | |
| "train_steps_per_second": 3.461 | |
| } | |
| ], | |
| "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": 1.5161929390646784e+18, | |
| "train_batch_size": 96, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |