{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 99.8648280558393, "eval_steps": 500, "global_step": 73400, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.9996595165134491, "grad_norm": 1.3896363973617554, "learning_rate": 9.90040871934605e-06, "loss": 2.3637, "step": 734 }, { "epoch": 1.9982975825672455, "grad_norm": 1.085627794265747, "learning_rate": 9.80040871934605e-06, "loss": 2.3048, "step": 1468 }, { "epoch": 2.996935648621042, "grad_norm": 1.7355504035949707, "learning_rate": 9.700544959128066e-06, "loss": 2.2626, "step": 2202 }, { "epoch": 3.9955737146748382, "grad_norm": 2.631619691848755, "learning_rate": 9.600544959128067e-06, "loss": 2.2366, "step": 2936 }, { "epoch": 4.994211780728635, "grad_norm": 3.0078556537628174, "learning_rate": 9.500544959128066e-06, "loss": 2.1902, "step": 3670 }, { "epoch": 5.992849846782431, "grad_norm": 3.808694362640381, "learning_rate": 9.400681198910083e-06, "loss": 2.1761, "step": 4404 }, { "epoch": 6.991487912836227, "grad_norm": 2.927497625350952, "learning_rate": 9.300681198910082e-06, "loss": 2.1423, "step": 5138 }, { "epoch": 7.990125978890024, "grad_norm": 3.1034231185913086, "learning_rate": 9.200681198910083e-06, "loss": 2.1311, "step": 5872 }, { "epoch": 8.98876404494382, "grad_norm": 2.8936243057250977, "learning_rate": 9.1008174386921e-06, "loss": 2.1006, "step": 6606 }, { "epoch": 9.987402110997616, "grad_norm": 3.1552979946136475, "learning_rate": 9.000817438692099e-06, "loss": 2.1108, "step": 7340 }, { "epoch": 10.986040177051413, "grad_norm": 4.212212562561035, "learning_rate": 8.900817438692098e-06, "loss": 2.0826, "step": 8074 }, { "epoch": 11.98467824310521, "grad_norm": 3.0641448497772217, "learning_rate": 8.800953678474115e-06, "loss": 2.0827, "step": 8808 }, { "epoch": 12.983316309159006, "grad_norm": 3.0309231281280518, "learning_rate": 8.700953678474116e-06, "loss": 2.0707, "step": 9542 }, { "epoch": 13.981954375212801, "grad_norm": 2.910205364227295, "learning_rate": 8.600953678474115e-06, "loss": 2.0545, "step": 10276 }, { "epoch": 14.980592441266598, "grad_norm": 2.4057796001434326, "learning_rate": 8.501089918256132e-06, "loss": 2.0518, "step": 11010 }, { "epoch": 15.979230507320395, "grad_norm": 3.38080096244812, "learning_rate": 8.401089918256131e-06, "loss": 2.0431, "step": 11744 }, { "epoch": 16.97786857337419, "grad_norm": 3.414673328399658, "learning_rate": 8.301089918256132e-06, "loss": 2.0278, "step": 12478 }, { "epoch": 17.97650663942799, "grad_norm": 3.2804346084594727, "learning_rate": 8.201089918256131e-06, "loss": 2.0394, "step": 13212 }, { "epoch": 18.975144705481785, "grad_norm": 3.313039541244507, "learning_rate": 8.101089918256132e-06, "loss": 2.0239, "step": 13946 }, { "epoch": 19.973782771535582, "grad_norm": 3.2474405765533447, "learning_rate": 8.001226158038149e-06, "loss": 2.0155, "step": 14680 }, { "epoch": 20.972420837589375, "grad_norm": 3.154122829437256, "learning_rate": 7.901226158038148e-06, "loss": 2.0183, "step": 15414 }, { "epoch": 21.971058903643172, "grad_norm": 4.2719407081604, "learning_rate": 7.801226158038147e-06, "loss": 2.0059, "step": 16148 }, { "epoch": 22.96969696969697, "grad_norm": 3.08585262298584, "learning_rate": 7.701226158038148e-06, "loss": 2.007, "step": 16882 }, { "epoch": 23.968335035750766, "grad_norm": 3.8600127696990967, "learning_rate": 7.601362397820165e-06, "loss": 1.9813, "step": 17616 }, { "epoch": 24.966973101804562, "grad_norm": 4.064390182495117, "learning_rate": 7.501362397820165e-06, "loss": 1.9909, "step": 18350 }, { "epoch": 25.96561116785836, "grad_norm": 5.1680169105529785, "learning_rate": 7.401362397820164e-06, "loss": 1.9781, "step": 19084 }, { "epoch": 26.964249233912156, "grad_norm": 3.716388702392578, "learning_rate": 7.30149863760218e-06, "loss": 1.9841, "step": 19818 }, { "epoch": 27.962887299965953, "grad_norm": 1.9636414051055908, "learning_rate": 7.20149863760218e-06, "loss": 1.9727, "step": 20552 }, { "epoch": 28.96152536601975, "grad_norm": 2.968393087387085, "learning_rate": 7.10149863760218e-06, "loss": 1.9729, "step": 21286 }, { "epoch": 29.960163432073543, "grad_norm": 1.9185447692871094, "learning_rate": 7.001498637602181e-06, "loss": 1.9586, "step": 22020 }, { "epoch": 30.95880149812734, "grad_norm": 3.3125815391540527, "learning_rate": 6.901634877384197e-06, "loss": 1.9661, "step": 22754 }, { "epoch": 31.957439564181136, "grad_norm": 3.3222153186798096, "learning_rate": 6.801634877384197e-06, "loss": 1.9532, "step": 23488 }, { "epoch": 32.95607763023494, "grad_norm": 2.3255455493927, "learning_rate": 6.701634877384197e-06, "loss": 1.9445, "step": 24222 }, { "epoch": 33.95471569628873, "grad_norm": 3.436879873275757, "learning_rate": 6.6017711171662135e-06, "loss": 1.9564, "step": 24956 }, { "epoch": 34.95335376234252, "grad_norm": 2.789512872695923, "learning_rate": 6.501771117166214e-06, "loss": 1.9458, "step": 25690 }, { "epoch": 35.951991828396324, "grad_norm": 2.496615409851074, "learning_rate": 6.401771117166214e-06, "loss": 1.9314, "step": 26424 }, { "epoch": 36.95062989445012, "grad_norm": 3.680979013442993, "learning_rate": 6.301907356948229e-06, "loss": 1.9481, "step": 27158 }, { "epoch": 37.94926796050392, "grad_norm": 3.8750393390655518, "learning_rate": 6.201907356948229e-06, "loss": 1.9346, "step": 27892 }, { "epoch": 38.94790602655771, "grad_norm": 3.358818531036377, "learning_rate": 6.101907356948229e-06, "loss": 1.9426, "step": 28626 }, { "epoch": 39.94654409261151, "grad_norm": 3.4985063076019287, "learning_rate": 6.001907356948229e-06, "loss": 1.9319, "step": 29360 }, { "epoch": 40.945182158665304, "grad_norm": 3.5801708698272705, "learning_rate": 5.9020435967302455e-06, "loss": 1.9316, "step": 30094 }, { "epoch": 41.943820224719104, "grad_norm": 2.131335973739624, "learning_rate": 5.802043596730246e-06, "loss": 1.9267, "step": 30828 }, { "epoch": 42.9424582907729, "grad_norm": 3.048663854598999, "learning_rate": 5.702043596730246e-06, "loss": 1.9313, "step": 31562 }, { "epoch": 43.94109635682669, "grad_norm": 4.18536376953125, "learning_rate": 5.602043596730246e-06, "loss": 1.9307, "step": 32296 }, { "epoch": 44.93973442288049, "grad_norm": 3.1449191570281982, "learning_rate": 5.502043596730245e-06, "loss": 1.9121, "step": 33030 }, { "epoch": 45.938372488934284, "grad_norm": 2.445446491241455, "learning_rate": 5.4021798365122625e-06, "loss": 1.9147, "step": 33764 }, { "epoch": 46.937010554988085, "grad_norm": 3.044128656387329, "learning_rate": 5.302316076294278e-06, "loss": 1.9179, "step": 34498 }, { "epoch": 47.93564862104188, "grad_norm": 2.411952018737793, "learning_rate": 5.202316076294278e-06, "loss": 1.9175, "step": 35232 }, { "epoch": 48.93428668709568, "grad_norm": 4.5417799949646, "learning_rate": 5.1023160762942784e-06, "loss": 1.9139, "step": 35966 }, { "epoch": 49.93292475314947, "grad_norm": 3.36789608001709, "learning_rate": 5.002452316076295e-06, "loss": 1.8999, "step": 36700 }, { "epoch": 50.93156281920327, "grad_norm": 1.4864888191223145, "learning_rate": 4.902452316076295e-06, "loss": 1.9151, "step": 37434 }, { "epoch": 51.930200885257065, "grad_norm": 2.543835163116455, "learning_rate": 4.802452316076295e-06, "loss": 1.9049, "step": 38168 }, { "epoch": 52.92883895131086, "grad_norm": 2.121729850769043, "learning_rate": 4.7024523160762945e-06, "loss": 1.9099, "step": 38902 }, { "epoch": 53.92747701736466, "grad_norm": 3.2268223762512207, "learning_rate": 4.602452316076295e-06, "loss": 1.9071, "step": 39636 }, { "epoch": 54.92611508341845, "grad_norm": 3.928720235824585, "learning_rate": 4.502452316076295e-06, "loss": 1.8883, "step": 40370 }, { "epoch": 55.92475314947225, "grad_norm": 2.6059372425079346, "learning_rate": 4.402588555858311e-06, "loss": 1.9033, "step": 41104 }, { "epoch": 56.923391215526046, "grad_norm": 2.899505615234375, "learning_rate": 4.3025885558583105e-06, "loss": 1.8852, "step": 41838 }, { "epoch": 57.922029281579846, "grad_norm": 2.1250553131103516, "learning_rate": 4.202588555858311e-06, "loss": 1.8919, "step": 42572 }, { "epoch": 58.92066734763364, "grad_norm": 3.6068265438079834, "learning_rate": 4.102588555858311e-06, "loss": 1.8801, "step": 43306 }, { "epoch": 59.91930541368744, "grad_norm": 3.1520488262176514, "learning_rate": 4.002588555858311e-06, "loss": 1.8997, "step": 44040 }, { "epoch": 60.91794347974123, "grad_norm": 2.7997710704803467, "learning_rate": 3.902588555858311e-06, "loss": 1.8942, "step": 44774 }, { "epoch": 61.916581545795026, "grad_norm": 2.2547781467437744, "learning_rate": 3.8027247956403276e-06, "loss": 1.8996, "step": 45508 }, { "epoch": 62.915219611848826, "grad_norm": 2.583367347717285, "learning_rate": 3.7027247956403272e-06, "loss": 1.8811, "step": 46242 }, { "epoch": 63.91385767790262, "grad_norm": 3.9086215496063232, "learning_rate": 3.6027247956403274e-06, "loss": 1.8878, "step": 46976 }, { "epoch": 64.91249574395641, "grad_norm": 1.8989681005477905, "learning_rate": 3.50299727520436e-06, "loss": 1.8987, "step": 47710 }, { "epoch": 65.91113381001021, "grad_norm": 2.5425479412078857, "learning_rate": 3.40299727520436e-06, "loss": 1.8928, "step": 48444 }, { "epoch": 66.90977187606401, "grad_norm": 2.8161380290985107, "learning_rate": 3.3029972752043598e-06, "loss": 1.8833, "step": 49178 }, { "epoch": 67.90840994211781, "grad_norm": 3.328853130340576, "learning_rate": 3.20299727520436e-06, "loss": 1.8802, "step": 49912 }, { "epoch": 68.9070480081716, "grad_norm": 1.836472749710083, "learning_rate": 3.1029972752043596e-06, "loss": 1.8821, "step": 50646 }, { "epoch": 69.9056860742254, "grad_norm": 3.308563470840454, "learning_rate": 3.00299727520436e-06, "loss": 1.8824, "step": 51380 }, { "epoch": 70.9043241402792, "grad_norm": 2.7526936531066895, "learning_rate": 2.9031335149863767e-06, "loss": 1.8828, "step": 52114 }, { "epoch": 71.90296220633299, "grad_norm": 2.601332664489746, "learning_rate": 2.8031335149863763e-06, "loss": 1.8872, "step": 52848 }, { "epoch": 72.90160027238679, "grad_norm": 3.038666248321533, "learning_rate": 2.7031335149863765e-06, "loss": 1.8861, "step": 53582 }, { "epoch": 73.90023833844059, "grad_norm": 2.3815577030181885, "learning_rate": 2.6032697547683926e-06, "loss": 1.8768, "step": 54316 }, { "epoch": 74.89887640449439, "grad_norm": 3.253783941268921, "learning_rate": 2.5032697547683927e-06, "loss": 1.8767, "step": 55050 }, { "epoch": 75.89751447054817, "grad_norm": 1.8898478746414185, "learning_rate": 2.403405994550409e-06, "loss": 1.8778, "step": 55784 }, { "epoch": 76.89615253660197, "grad_norm": 2.8832762241363525, "learning_rate": 2.303405994550409e-06, "loss": 1.8816, "step": 56518 }, { "epoch": 77.89479060265577, "grad_norm": 2.767455577850342, "learning_rate": 2.2034059945504087e-06, "loss": 1.8817, "step": 57252 }, { "epoch": 78.89342866870956, "grad_norm": 1.9658331871032715, "learning_rate": 2.1035422343324252e-06, "loss": 1.8857, "step": 57986 }, { "epoch": 79.89206673476336, "grad_norm": 2.7493135929107666, "learning_rate": 2.003542234332425e-06, "loss": 1.8763, "step": 58720 }, { "epoch": 80.89070480081716, "grad_norm": 2.519388198852539, "learning_rate": 1.9035422343324252e-06, "loss": 1.8782, "step": 59454 }, { "epoch": 81.88934286687096, "grad_norm": 2.258284091949463, "learning_rate": 1.8036784741144416e-06, "loss": 1.8709, "step": 60188 }, { "epoch": 82.88798093292475, "grad_norm": 2.6191112995147705, "learning_rate": 1.7038147138964577e-06, "loss": 1.8486, "step": 60922 }, { "epoch": 83.88661899897855, "grad_norm": 3.241089105606079, "learning_rate": 1.603814713896458e-06, "loss": 1.8726, "step": 61656 }, { "epoch": 84.88525706503235, "grad_norm": 1.7192323207855225, "learning_rate": 1.503814713896458e-06, "loss": 1.8793, "step": 62390 }, { "epoch": 85.88389513108615, "grad_norm": 2.782193899154663, "learning_rate": 1.4038147138964579e-06, "loss": 1.8714, "step": 63124 }, { "epoch": 86.88253319713994, "grad_norm": 2.659689426422119, "learning_rate": 1.303814713896458e-06, "loss": 1.8908, "step": 63858 }, { "epoch": 87.88117126319374, "grad_norm": 1.9582741260528564, "learning_rate": 1.2038147138964579e-06, "loss": 1.8595, "step": 64592 }, { "epoch": 88.87980932924754, "grad_norm": 1.9837833642959595, "learning_rate": 1.1039509536784742e-06, "loss": 1.875, "step": 65326 }, { "epoch": 89.87844739530132, "grad_norm": 2.076040506362915, "learning_rate": 1.0039509536784741e-06, "loss": 1.8664, "step": 66060 }, { "epoch": 90.87708546135512, "grad_norm": 1.5430365800857544, "learning_rate": 9.039509536784742e-07, "loss": 1.8875, "step": 66794 }, { "epoch": 91.87572352740892, "grad_norm": 1.8234200477600098, "learning_rate": 8.040871934604905e-07, "loss": 1.868, "step": 67528 }, { "epoch": 92.87436159346272, "grad_norm": 2.1992905139923096, "learning_rate": 7.040871934604905e-07, "loss": 1.8844, "step": 68262 }, { "epoch": 93.87299965951651, "grad_norm": 2.1095755100250244, "learning_rate": 6.042234332425069e-07, "loss": 1.8808, "step": 68996 }, { "epoch": 94.87163772557031, "grad_norm": 2.1402716636657715, "learning_rate": 5.042234332425069e-07, "loss": 1.8661, "step": 69730 }, { "epoch": 95.87027579162411, "grad_norm": 2.9710769653320312, "learning_rate": 4.0422343324250687e-07, "loss": 1.8651, "step": 70464 }, { "epoch": 96.8689138576779, "grad_norm": 1.769300103187561, "learning_rate": 3.043596730245232e-07, "loss": 1.8726, "step": 71198 }, { "epoch": 97.8675519237317, "grad_norm": 2.095330238342285, "learning_rate": 2.0435967302452318e-07, "loss": 1.8691, "step": 71932 }, { "epoch": 98.8661899897855, "grad_norm": 2.741286039352417, "learning_rate": 1.0435967302452316e-07, "loss": 1.8809, "step": 72666 }, { "epoch": 99.8648280558393, "grad_norm": 2.270244836807251, "learning_rate": 4.49591280653951e-09, "loss": 1.8619, "step": 73400 }, { "epoch": 99.8648280558393, "step": 73400, "total_flos": 3.848614248660486e+18, "train_loss": 1.9516082713779377, "train_runtime": 79711.1977, "train_samples_per_second": 184.18, "train_steps_per_second": 0.921 } ], "logging_steps": 734, "max_steps": 73400, "num_input_tokens_seen": 0, "num_train_epochs": 100, "save_steps": 6800, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 3.848614248660486e+18, "train_batch_size": 50, "trial_name": null, "trial_params": null }