| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 1000, | |
| "global_step": 8503, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005880277549100318, | |
| "grad_norm": 387.39825439453125, | |
| "learning_rate": 5.875440658049355e-07, | |
| "loss": 5.0951, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.011760555098200636, | |
| "grad_norm": 66.9789047241211, | |
| "learning_rate": 1.175088131609871e-06, | |
| "loss": 0.8587, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.01764083264730095, | |
| "grad_norm": 11.23346996307373, | |
| "learning_rate": 1.762632197414806e-06, | |
| "loss": 0.084, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.02352111019640127, | |
| "grad_norm": 11.451313018798828, | |
| "learning_rate": 2.350176263219742e-06, | |
| "loss": 0.051, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.02940138774550159, | |
| "grad_norm": 12.508331298828125, | |
| "learning_rate": 2.937720329024677e-06, | |
| "loss": 0.0592, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.0352816652946019, | |
| "grad_norm": 42.7663688659668, | |
| "learning_rate": 3.525264394829612e-06, | |
| "loss": 0.0519, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.041161942843702226, | |
| "grad_norm": 9.149981498718262, | |
| "learning_rate": 4.112808460634548e-06, | |
| "loss": 0.0418, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.04704222039280254, | |
| "grad_norm": 16.077157974243164, | |
| "learning_rate": 4.700352526439484e-06, | |
| "loss": 0.0665, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.05292249794190286, | |
| "grad_norm": 8.89875316619873, | |
| "learning_rate": 5.287896592244419e-06, | |
| "loss": 0.052, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.05880277549100318, | |
| "grad_norm": 10.934527397155762, | |
| "learning_rate": 5.875440658049354e-06, | |
| "loss": 0.0405, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.0646830530401035, | |
| "grad_norm": 5.134288787841797, | |
| "learning_rate": 6.46298472385429e-06, | |
| "loss": 0.0499, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.0705633305892038, | |
| "grad_norm": 2.750032663345337, | |
| "learning_rate": 7.050528789659224e-06, | |
| "loss": 0.0444, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.07644360813830413, | |
| "grad_norm": 3.613896131515503, | |
| "learning_rate": 7.63807285546416e-06, | |
| "loss": 0.0417, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.08232388568740445, | |
| "grad_norm": 5.510785102844238, | |
| "learning_rate": 8.225616921269097e-06, | |
| "loss": 0.0337, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.08820416323650476, | |
| "grad_norm": 21.227962493896484, | |
| "learning_rate": 8.81316098707403e-06, | |
| "loss": 0.0314, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.09408444078560509, | |
| "grad_norm": 11.67798137664795, | |
| "learning_rate": 9.400705052878968e-06, | |
| "loss": 0.0386, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.0999647183347054, | |
| "grad_norm": 0.9770076870918274, | |
| "learning_rate": 9.988249118683903e-06, | |
| "loss": 0.0322, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.10584499588380572, | |
| "grad_norm": 2.259112596511841, | |
| "learning_rate": 9.998988263671598e-06, | |
| "loss": 0.037, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.11172527343290603, | |
| "grad_norm": 5.629507064819336, | |
| "learning_rate": 9.995870471854679e-06, | |
| "loss": 0.0328, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.11760555098200635, | |
| "grad_norm": 2.3545737266540527, | |
| "learning_rate": 9.990647516930925e-06, | |
| "loss": 0.0331, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11760555098200635, | |
| "eval_loss": 0.03788302466273308, | |
| "eval_runtime": 619.4182, | |
| "eval_samples_per_second": 47.064, | |
| "eval_steps_per_second": 11.766, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.12348582853110666, | |
| "grad_norm": 1.4001481533050537, | |
| "learning_rate": 9.983321599752438e-06, | |
| "loss": 0.0359, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.129366106080207, | |
| "grad_norm": 10.03685188293457, | |
| "learning_rate": 9.9738958073189e-06, | |
| "loss": 0.0306, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.1352463836293073, | |
| "grad_norm": 8.22728443145752, | |
| "learning_rate": 9.962374111476778e-06, | |
| "loss": 0.031, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.1411266611784076, | |
| "grad_norm": 4.021151065826416, | |
| "learning_rate": 9.948761367245665e-06, | |
| "loss": 0.0333, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.14700693872750795, | |
| "grad_norm": 4.213313102722168, | |
| "learning_rate": 9.933063310772463e-06, | |
| "loss": 0.0347, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.15288721627660826, | |
| "grad_norm": 1.7054405212402344, | |
| "learning_rate": 9.915286556914286e-06, | |
| "loss": 0.0326, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.15876749382570857, | |
| "grad_norm": 3.276946783065796, | |
| "learning_rate": 9.89543859645109e-06, | |
| "loss": 0.0323, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.1646477713748089, | |
| "grad_norm": 5.019853591918945, | |
| "learning_rate": 9.873527792929196e-06, | |
| "loss": 0.0274, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.17052804892390921, | |
| "grad_norm": 6.062413692474365, | |
| "learning_rate": 9.84956337913706e-06, | |
| "loss": 0.0294, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.17640832647300952, | |
| "grad_norm": 4.063215255737305, | |
| "learning_rate": 9.82355545321475e-06, | |
| "loss": 0.0336, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18228860402210983, | |
| "grad_norm": 3.2010011672973633, | |
| "learning_rate": 9.795514974398789e-06, | |
| "loss": 0.0295, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.18816888157121017, | |
| "grad_norm": 4.517515182495117, | |
| "learning_rate": 9.765453758404144e-06, | |
| "loss": 0.0301, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.19404915912031048, | |
| "grad_norm": 4.797945022583008, | |
| "learning_rate": 9.733384472445308e-06, | |
| "loss": 0.0309, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.1999294366694108, | |
| "grad_norm": 2.705932378768921, | |
| "learning_rate": 9.699320629898589e-06, | |
| "loss": 0.026, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.2058097142185111, | |
| "grad_norm": 20.035690307617188, | |
| "learning_rate": 9.663276584607831e-06, | |
| "loss": 0.0264, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.21168999176761144, | |
| "grad_norm": 1.2530241012573242, | |
| "learning_rate": 9.625267524835974e-06, | |
| "loss": 0.0277, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.21757026931671175, | |
| "grad_norm": 3.373883008956909, | |
| "learning_rate": 9.585309466865029e-06, | |
| "loss": 0.0276, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.22345054686581206, | |
| "grad_norm": 5.847758769989014, | |
| "learning_rate": 9.54341924824712e-06, | |
| "loss": 0.0289, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.22933082441491237, | |
| "grad_norm": 1.0913113355636597, | |
| "learning_rate": 9.499614520709457e-06, | |
| "loss": 0.0256, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.2352111019640127, | |
| "grad_norm": 4.371528148651123, | |
| "learning_rate": 9.453913742716256e-06, | |
| "loss": 0.0299, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2352111019640127, | |
| "eval_loss": 0.026240274310112, | |
| "eval_runtime": 619.3143, | |
| "eval_samples_per_second": 47.071, | |
| "eval_steps_per_second": 11.768, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.24109137951311302, | |
| "grad_norm": 4.8858208656311035, | |
| "learning_rate": 9.40633617169069e-06, | |
| "loss": 0.0273, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.24697165706221333, | |
| "grad_norm": 1.7557135820388794, | |
| "learning_rate": 9.35690185590018e-06, | |
| "loss": 0.0265, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.25285193461131367, | |
| "grad_norm": 2.1983985900878906, | |
| "learning_rate": 9.305631626008454e-06, | |
| "loss": 0.0249, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.258732212160414, | |
| "grad_norm": 4.288529872894287, | |
| "learning_rate": 9.252547086297895e-06, | |
| "loss": 0.0243, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.2646124897095143, | |
| "grad_norm": 1.228149652481079, | |
| "learning_rate": 9.197670605565932e-06, | |
| "loss": 0.0271, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.2704927672586146, | |
| "grad_norm": 4.099296569824219, | |
| "learning_rate": 9.141025307699246e-06, | |
| "loss": 0.0265, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.2763730448077149, | |
| "grad_norm": 3.512291431427002, | |
| "learning_rate": 9.082635061929817e-06, | |
| "loss": 0.0286, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.2822533223568152, | |
| "grad_norm": 1.2961852550506592, | |
| "learning_rate": 9.022524472776897e-06, | |
| "loss": 0.0288, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.2881335999059156, | |
| "grad_norm": 4.3626322746276855, | |
| "learning_rate": 8.960718869679132e-06, | |
| "loss": 0.0252, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.2940138774550159, | |
| "grad_norm": 1.3140321969985962, | |
| "learning_rate": 8.89724429632124e-06, | |
| "loss": 0.0243, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.2998941550041162, | |
| "grad_norm": 4.2938432693481445, | |
| "learning_rate": 8.832127499659687e-06, | |
| "loss": 0.0216, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.3057744325532165, | |
| "grad_norm": 2.0306389331817627, | |
| "learning_rate": 8.765395918652062e-06, | |
| "loss": 0.0245, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.3116547101023168, | |
| "grad_norm": 3.160212993621826, | |
| "learning_rate": 8.697077672694809e-06, | |
| "loss": 0.0292, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.31753498765141713, | |
| "grad_norm": 8.705582618713379, | |
| "learning_rate": 8.627201549774273e-06, | |
| "loss": 0.0243, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.32341526520051744, | |
| "grad_norm": 5.516971588134766, | |
| "learning_rate": 8.55579699433599e-06, | |
| "loss": 0.0272, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.3292955427496178, | |
| "grad_norm": 3.2388885021209717, | |
| "learning_rate": 8.482894094877372e-06, | |
| "loss": 0.0268, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.3351758202987181, | |
| "grad_norm": 1.1581066846847534, | |
| "learning_rate": 8.408523571269e-06, | |
| "loss": 0.0251, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.34105609784781843, | |
| "grad_norm": 1.0647870302200317, | |
| "learning_rate": 8.332716761809857e-06, | |
| "loss": 0.0304, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.34693637539691874, | |
| "grad_norm": 1.6037498712539673, | |
| "learning_rate": 8.255505610021981e-06, | |
| "loss": 0.0241, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.35281665294601905, | |
| "grad_norm": 3.1624505519866943, | |
| "learning_rate": 8.176922651190085e-06, | |
| "loss": 0.0251, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.35281665294601905, | |
| "eval_loss": 0.028408875688910484, | |
| "eval_runtime": 619.7012, | |
| "eval_samples_per_second": 47.042, | |
| "eval_steps_per_second": 11.761, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.35869693049511936, | |
| "grad_norm": 4.156674861907959, | |
| "learning_rate": 8.097000998651812e-06, | |
| "loss": 0.0316, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.36457720804421967, | |
| "grad_norm": 1.2711337804794312, | |
| "learning_rate": 8.015774329844417e-06, | |
| "loss": 0.0239, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.37045748559332, | |
| "grad_norm": 1.4552150964736938, | |
| "learning_rate": 7.933276872113754e-06, | |
| "loss": 0.0226, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.37633776314242035, | |
| "grad_norm": 6.432965278625488, | |
| "learning_rate": 7.849543388291524e-06, | |
| "loss": 0.0256, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.38221804069152066, | |
| "grad_norm": 3.6660749912261963, | |
| "learning_rate": 7.764609162046894e-06, | |
| "loss": 0.0265, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.38809831824062097, | |
| "grad_norm": 3.6166951656341553, | |
| "learning_rate": 7.678509983018656e-06, | |
| "loss": 0.0256, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.3939785957897213, | |
| "grad_norm": 8.923437118530273, | |
| "learning_rate": 7.591282131734139e-06, | |
| "loss": 0.024, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 0.3998588733388216, | |
| "grad_norm": 3.535114049911499, | |
| "learning_rate": 7.50296236432132e-06, | |
| "loss": 0.0208, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.4057391508879219, | |
| "grad_norm": 3.5117030143737793, | |
| "learning_rate": 7.413587897020496e-06, | |
| "loss": 0.0243, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 0.4116194284370222, | |
| "grad_norm": 1.3753975629806519, | |
| "learning_rate": 7.323196390502074e-06, | |
| "loss": 0.0213, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.41749970598612257, | |
| "grad_norm": 1.682301640510559, | |
| "learning_rate": 7.231825933997105e-06, | |
| "loss": 0.0261, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 0.4233799835352229, | |
| "grad_norm": 4.877748489379883, | |
| "learning_rate": 7.139515029247213e-06, | |
| "loss": 0.0286, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.4292602610843232, | |
| "grad_norm": 1.995045781135559, | |
| "learning_rate": 7.046302574280703e-06, | |
| "loss": 0.0244, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 0.4351405386334235, | |
| "grad_norm": 2.8061089515686035, | |
| "learning_rate": 6.952227847021697e-06, | |
| "loss": 0.023, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.4410208161825238, | |
| "grad_norm": 1.9934951066970825, | |
| "learning_rate": 6.857330488739159e-06, | |
| "loss": 0.0224, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.4469010937316241, | |
| "grad_norm": 4.9554290771484375, | |
| "learning_rate": 6.76165048734285e-06, | |
| "loss": 0.0235, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.45278137128072443, | |
| "grad_norm": 2.459325075149536, | |
| "learning_rate": 6.665228160533186e-06, | |
| "loss": 0.0251, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 0.45866164882982474, | |
| "grad_norm": 3.0527613162994385, | |
| "learning_rate": 6.568104138812141e-06, | |
| "loss": 0.0208, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.4645419263789251, | |
| "grad_norm": 2.7639405727386475, | |
| "learning_rate": 6.470319348362344e-06, | |
| "loss": 0.0242, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 0.4704222039280254, | |
| "grad_norm": 1.0516586303710938, | |
| "learning_rate": 6.371914993801573e-06, | |
| "loss": 0.0213, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4704222039280254, | |
| "eval_loss": 0.02518468163907528, | |
| "eval_runtime": 612.2227, | |
| "eval_samples_per_second": 47.617, | |
| "eval_steps_per_second": 11.904, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4763024814771257, | |
| "grad_norm": 2.6117072105407715, | |
| "learning_rate": 6.272932540819929e-06, | |
| "loss": 0.0201, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 0.48218275902622604, | |
| "grad_norm": 3.2772061824798584, | |
| "learning_rate": 6.173413698706999e-06, | |
| "loss": 0.0245, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.48806303657532635, | |
| "grad_norm": 9.428531646728516, | |
| "learning_rate": 6.073400402776364e-06, | |
| "loss": 0.0229, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 0.49394331412442666, | |
| "grad_norm": 3.652819871902466, | |
| "learning_rate": 5.972934796694871e-06, | |
| "loss": 0.0203, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.49982359167352697, | |
| "grad_norm": 3.571290969848633, | |
| "learning_rate": 5.872059214724112e-06, | |
| "loss": 0.0224, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.5057038692226273, | |
| "grad_norm": 7.393253326416016, | |
| "learning_rate": 5.770816163881581e-06, | |
| "loss": 0.0233, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.5115841467717276, | |
| "grad_norm": 3.7029826641082764, | |
| "learning_rate": 5.669248306029042e-06, | |
| "loss": 0.0211, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 0.517464424320828, | |
| "grad_norm": 1.5849096775054932, | |
| "learning_rate": 5.567398439895643e-06, | |
| "loss": 0.0228, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.5233447018699282, | |
| "grad_norm": 4.666280269622803, | |
| "learning_rate": 5.465309483043364e-06, | |
| "loss": 0.019, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 0.5292249794190286, | |
| "grad_norm": 1.970779299736023, | |
| "learning_rate": 5.363024453782388e-06, | |
| "loss": 0.0262, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.5351052569681289, | |
| "grad_norm": 3.114126682281494, | |
| "learning_rate": 5.260586453044011e-06, | |
| "loss": 0.0208, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 0.5409855345172292, | |
| "grad_norm": 2.33022141456604, | |
| "learning_rate": 5.158038646218749e-06, | |
| "loss": 0.0248, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.5468658120663296, | |
| "grad_norm": 1.2761564254760742, | |
| "learning_rate": 5.055424244967284e-06, | |
| "loss": 0.0206, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 0.5527460896154298, | |
| "grad_norm": 1.679206132888794, | |
| "learning_rate": 4.95278648901189e-06, | |
| "loss": 0.0202, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.5586263671645302, | |
| "grad_norm": 3.895859479904175, | |
| "learning_rate": 4.850168627916068e-06, | |
| "loss": 0.023, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.5645066447136304, | |
| "grad_norm": 1.8703380823135376, | |
| "learning_rate": 4.7476139028600085e-06, | |
| "loss": 0.0243, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.5703869222627308, | |
| "grad_norm": 1.986175298690796, | |
| "learning_rate": 4.645165528419598e-06, | |
| "loss": 0.0199, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 0.5762671998118312, | |
| "grad_norm": 1.6648274660110474, | |
| "learning_rate": 4.542866674356627e-06, | |
| "loss": 0.0224, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.5821474773609314, | |
| "grad_norm": 1.7521028518676758, | |
| "learning_rate": 4.440760447427899e-06, | |
| "loss": 0.0206, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 0.5880277549100318, | |
| "grad_norm": 3.9067554473876953, | |
| "learning_rate": 4.338889873220875e-06, | |
| "loss": 0.0264, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5880277549100318, | |
| "eval_loss": 0.022188851609826088, | |
| "eval_runtime": 615.8241, | |
| "eval_samples_per_second": 47.338, | |
| "eval_steps_per_second": 11.835, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.593908032459132, | |
| "grad_norm": 2.8073413372039795, | |
| "learning_rate": 4.237297878023512e-06, | |
| "loss": 0.0238, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 0.5997883100082324, | |
| "grad_norm": 2.105537176132202, | |
| "learning_rate": 4.136027270735971e-06, | |
| "loss": 0.0207, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.6056685875573327, | |
| "grad_norm": 3.5026049613952637, | |
| "learning_rate": 4.035120724831766e-06, | |
| "loss": 0.0208, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 0.611548865106433, | |
| "grad_norm": 3.762202262878418, | |
| "learning_rate": 3.9346207603759966e-06, | |
| "loss": 0.0214, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.6174291426555334, | |
| "grad_norm": 2.426508903503418, | |
| "learning_rate": 3.834569726108201e-06, | |
| "loss": 0.0195, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 0.6233094202046336, | |
| "grad_norm": 2.668912410736084, | |
| "learning_rate": 3.7350097815974395e-06, | |
| "loss": 0.0229, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.629189697753734, | |
| "grad_norm": 3.8856732845306396, | |
| "learning_rate": 3.6359828794770467e-06, | |
| "loss": 0.0253, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 0.6350699753028343, | |
| "grad_norm": 0.85421222448349, | |
| "learning_rate": 3.5375307477666134e-06, | |
| "loss": 0.0197, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.6409502528519346, | |
| "grad_norm": 6.5461225509643555, | |
| "learning_rate": 3.4396948722886065e-06, | |
| "loss": 0.018, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 0.6468305304010349, | |
| "grad_norm": 4.219273567199707, | |
| "learning_rate": 3.342516479187047e-06, | |
| "loss": 0.019, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.6527108079501353, | |
| "grad_norm": 2.8508059978485107, | |
| "learning_rate": 3.246036517555611e-06, | |
| "loss": 0.0218, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 0.6585910854992356, | |
| "grad_norm": 0.25915786623954773, | |
| "learning_rate": 3.1502956421824714e-06, | |
| "loss": 0.0172, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.6644713630483359, | |
| "grad_norm": 3.543858766555786, | |
| "learning_rate": 3.0553341964191587e-06, | |
| "loss": 0.0213, | |
| "step": 5650 | |
| }, | |
| { | |
| "epoch": 0.6703516405974362, | |
| "grad_norm": 4.268918514251709, | |
| "learning_rate": 2.961192195180657e-06, | |
| "loss": 0.0175, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 0.6762319181465365, | |
| "grad_norm": 3.376847982406616, | |
| "learning_rate": 2.867909308083885e-06, | |
| "loss": 0.0167, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 0.6821121956956369, | |
| "grad_norm": 3.2170565128326416, | |
| "learning_rate": 2.7755248427316976e-06, | |
| "loss": 0.0192, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.6879924732447371, | |
| "grad_norm": 4.281452178955078, | |
| "learning_rate": 2.68407772814942e-06, | |
| "loss": 0.0176, | |
| "step": 5850 | |
| }, | |
| { | |
| "epoch": 0.6938727507938375, | |
| "grad_norm": 2.3371615409851074, | |
| "learning_rate": 2.5936064983808994e-06, | |
| "loss": 0.0212, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 0.6997530283429377, | |
| "grad_norm": 3.5211522579193115, | |
| "learning_rate": 2.5041492762510245e-06, | |
| "loss": 0.0206, | |
| "step": 5950 | |
| }, | |
| { | |
| "epoch": 0.7056333058920381, | |
| "grad_norm": 3.2257282733917236, | |
| "learning_rate": 2.415743757301486e-06, | |
| "loss": 0.0183, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.7056333058920381, | |
| "eval_loss": 0.019069144502282143, | |
| "eval_runtime": 622.7936, | |
| "eval_samples_per_second": 46.808, | |
| "eval_steps_per_second": 11.702, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.7115135834411385, | |
| "grad_norm": 1.6855459213256836, | |
| "learning_rate": 2.3284271939066127e-06, | |
| "loss": 0.0175, | |
| "step": 6050 | |
| }, | |
| { | |
| "epoch": 0.7173938609902387, | |
| "grad_norm": 1.7634873390197754, | |
| "learning_rate": 2.2422363795759534e-06, | |
| "loss": 0.0212, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 0.7232741385393391, | |
| "grad_norm": 4.85552978515625, | |
| "learning_rate": 2.157207633450183e-06, | |
| "loss": 0.0192, | |
| "step": 6150 | |
| }, | |
| { | |
| "epoch": 0.7291544160884393, | |
| "grad_norm": 1.8883658647537231, | |
| "learning_rate": 2.073376784996931e-06, | |
| "loss": 0.0205, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.7350346936375397, | |
| "grad_norm": 2.118055820465088, | |
| "learning_rate": 1.990779158912943e-06, | |
| "loss": 0.0179, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 0.74091497118664, | |
| "grad_norm": 0.919160783290863, | |
| "learning_rate": 1.9094495602389235e-06, | |
| "loss": 0.0172, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 0.7467952487357403, | |
| "grad_norm": 1.7510900497436523, | |
| "learning_rate": 1.829422259693377e-06, | |
| "loss": 0.0186, | |
| "step": 6350 | |
| }, | |
| { | |
| "epoch": 0.7526755262848407, | |
| "grad_norm": 3.255988597869873, | |
| "learning_rate": 1.750730979231588e-06, | |
| "loss": 0.0209, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.758555803833941, | |
| "grad_norm": 4.849563121795654, | |
| "learning_rate": 1.6734088778358371e-06, | |
| "loss": 0.0217, | |
| "step": 6450 | |
| }, | |
| { | |
| "epoch": 0.7644360813830413, | |
| "grad_norm": 2.414062261581421, | |
| "learning_rate": 1.5974885375428494e-06, | |
| "loss": 0.0169, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.7703163589321416, | |
| "grad_norm": 2.1367931365966797, | |
| "learning_rate": 1.5230019497143633e-06, | |
| "loss": 0.0207, | |
| "step": 6550 | |
| }, | |
| { | |
| "epoch": 0.7761966364812419, | |
| "grad_norm": 5.911067008972168, | |
| "learning_rate": 1.4499805015565754e-06, | |
| "loss": 0.0158, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.7820769140303422, | |
| "grad_norm": 2.751194953918457, | |
| "learning_rate": 1.378454962894193e-06, | |
| "loss": 0.0189, | |
| "step": 6650 | |
| }, | |
| { | |
| "epoch": 0.7879571915794426, | |
| "grad_norm": 2.061358690261841, | |
| "learning_rate": 1.308455473204619e-06, | |
| "loss": 0.0171, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 0.7938374691285429, | |
| "grad_norm": 2.2449493408203125, | |
| "learning_rate": 1.240011528917756e-06, | |
| "loss": 0.019, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 0.7997177466776432, | |
| "grad_norm": 4.053592205047607, | |
| "learning_rate": 1.1731519709867933e-06, | |
| "loss": 0.0167, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.8055980242267435, | |
| "grad_norm": 0.6525304317474365, | |
| "learning_rate": 1.1079049727351726e-06, | |
| "loss": 0.0184, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 0.8114783017758438, | |
| "grad_norm": 1.0814751386642456, | |
| "learning_rate": 1.0442980279849086e-06, | |
| "loss": 0.0193, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 0.8173585793249442, | |
| "grad_norm": 4.694520473480225, | |
| "learning_rate": 9.823579394712175e-07, | |
| "loss": 0.0195, | |
| "step": 6950 | |
| }, | |
| { | |
| "epoch": 0.8232388568740444, | |
| "grad_norm": 3.2695419788360596, | |
| "learning_rate": 9.221108075483615e-07, | |
| "loss": 0.0171, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.8232388568740444, | |
| "eval_loss": 0.017895469442009926, | |
| "eval_runtime": 618.1531, | |
| "eval_samples_per_second": 47.16, | |
| "eval_steps_per_second": 11.79, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.8291191344231448, | |
| "grad_norm": 2.8549704551696777, | |
| "learning_rate": 8.63582019191469e-07, | |
| "loss": 0.0191, | |
| "step": 7050 | |
| }, | |
| { | |
| "epoch": 0.8349994119722451, | |
| "grad_norm": 1.8346871137619019, | |
| "learning_rate": 8.067962372989563e-07, | |
| "loss": 0.0168, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 0.8408796895213454, | |
| "grad_norm": 1.3449316024780273, | |
| "learning_rate": 7.517773903000519e-07, | |
| "loss": 0.0199, | |
| "step": 7150 | |
| }, | |
| { | |
| "epoch": 0.8467599670704458, | |
| "grad_norm": 1.1152535676956177, | |
| "learning_rate": 6.98548662071828e-07, | |
| "loss": 0.0166, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.852640244619546, | |
| "grad_norm": 2.3504021167755127, | |
| "learning_rate": 6.471324821699603e-07, | |
| "loss": 0.0156, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 0.8585205221686464, | |
| "grad_norm": 0.3554767072200775, | |
| "learning_rate": 5.975505163773437e-07, | |
| "loss": 0.0215, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 0.8644007997177466, | |
| "grad_norm": 0.13760164380073547, | |
| "learning_rate": 5.498236575745564e-07, | |
| "loss": 0.0196, | |
| "step": 7350 | |
| }, | |
| { | |
| "epoch": 0.870281077266847, | |
| "grad_norm": 1.4082413911819458, | |
| "learning_rate": 5.039720169360013e-07, | |
| "loss": 0.0159, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.8761613548159473, | |
| "grad_norm": 3.8390660285949707, | |
| "learning_rate": 4.600149154554501e-07, | |
| "loss": 0.0214, | |
| "step": 7450 | |
| }, | |
| { | |
| "epoch": 0.8820416323650476, | |
| "grad_norm": 3.1669695377349854, | |
| "learning_rate": 4.179708758045431e-07, | |
| "loss": 0.0181, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.887921909914148, | |
| "grad_norm": 6.687104225158691, | |
| "learning_rate": 3.7785761452770295e-07, | |
| "loss": 0.02, | |
| "step": 7550 | |
| }, | |
| { | |
| "epoch": 0.8938021874632482, | |
| "grad_norm": 4.694364070892334, | |
| "learning_rate": 3.396920345767185e-07, | |
| "loss": 0.0203, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.8996824650123486, | |
| "grad_norm": 2.0906476974487305, | |
| "learning_rate": 3.0349021818817326e-07, | |
| "loss": 0.0176, | |
| "step": 7650 | |
| }, | |
| { | |
| "epoch": 0.9055627425614489, | |
| "grad_norm": 3.2009048461914062, | |
| "learning_rate": 2.692674201066975e-07, | |
| "loss": 0.0189, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 0.9114430201105492, | |
| "grad_norm": 4.380706310272217, | |
| "learning_rate": 2.3703806115691951e-07, | |
| "loss": 0.0171, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 0.9173232976596495, | |
| "grad_norm": 0.9385294318199158, | |
| "learning_rate": 2.068157221668049e-07, | |
| "loss": 0.0161, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.9232035752087498, | |
| "grad_norm": 2.8777430057525635, | |
| "learning_rate": 1.786131382449602e-07, | |
| "loss": 0.0232, | |
| "step": 7850 | |
| }, | |
| { | |
| "epoch": 0.9290838527578502, | |
| "grad_norm": 0.5679008960723877, | |
| "learning_rate": 1.5244219341430443e-07, | |
| "loss": 0.0197, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 0.9349641303069505, | |
| "grad_norm": 2.9544992446899414, | |
| "learning_rate": 1.2831391560437278e-07, | |
| "loss": 0.0187, | |
| "step": 7950 | |
| }, | |
| { | |
| "epoch": 0.9408444078560508, | |
| "grad_norm": 2.460313081741333, | |
| "learning_rate": 1.0623847200435966e-07, | |
| "loss": 0.0185, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.9408444078560508, | |
| "eval_loss": 0.01769772544503212, | |
| "eval_runtime": 613.4256, | |
| "eval_samples_per_second": 47.523, | |
| "eval_steps_per_second": 11.881, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.9467246854051511, | |
| "grad_norm": 3.39328670501709, | |
| "learning_rate": 8.62251647788609e-08, | |
| "loss": 0.019, | |
| "step": 8050 | |
| }, | |
| { | |
| "epoch": 0.9526049629542515, | |
| "grad_norm": 4.006555080413818, | |
| "learning_rate": 6.828242714812527e-08, | |
| "loss": 0.0183, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 0.9584852405033517, | |
| "grad_norm": 2.6548595428466797, | |
| "learning_rate": 5.2417819834454374e-08, | |
| "loss": 0.0156, | |
| "step": 8150 | |
| }, | |
| { | |
| "epoch": 0.9643655180524521, | |
| "grad_norm": 1.8154182434082031, | |
| "learning_rate": 3.863802787626325e-08, | |
| "loss": 0.0175, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.9702457956015524, | |
| "grad_norm": 4.321004867553711, | |
| "learning_rate": 2.694885781113432e-08, | |
| "loss": 0.0195, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 0.9761260731506527, | |
| "grad_norm": 2.2515978813171387, | |
| "learning_rate": 1.735523522905347e-08, | |
| "loss": 0.0182, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 0.9820063506997531, | |
| "grad_norm": 2.5015530586242676, | |
| "learning_rate": 9.861202696864191e-09, | |
| "loss": 0.0197, | |
| "step": 8350 | |
| }, | |
| { | |
| "epoch": 0.9878866282488533, | |
| "grad_norm": 2.121727705001831, | |
| "learning_rate": 4.469918054806344e-09, | |
| "loss": 0.0156, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.9937669057979537, | |
| "grad_norm": 3.144368886947632, | |
| "learning_rate": 1.1836530858633234e-09, | |
| "loss": 0.0197, | |
| "step": 8450 | |
| }, | |
| { | |
| "epoch": 0.9996471833470539, | |
| "grad_norm": 2.387354850769043, | |
| "learning_rate": 3.792558477266894e-12, | |
| "loss": 0.0191, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 8503, | |
| "total_flos": 1.8987963054248755e+17, | |
| "train_loss": 0.0597800020030789, | |
| "train_runtime": 12067.3267, | |
| "train_samples_per_second": 5.637, | |
| "train_steps_per_second": 0.705 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 8503, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.8987963054248755e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |