| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 184080, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04074315514993481, | |
| "grad_norm": 1.712476134300232, | |
| "learning_rate": 0.00029918513689700127, | |
| "loss": 1.0726, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.08148631029986962, | |
| "grad_norm": 1.8538917303085327, | |
| "learning_rate": 0.0002983702737940026, | |
| "loss": 0.8568, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.12222946544980444, | |
| "grad_norm": 3.2072715759277344, | |
| "learning_rate": 0.0002975554106910039, | |
| "loss": 0.7747, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.16297262059973924, | |
| "grad_norm": 2.7058048248291016, | |
| "learning_rate": 0.0002967405475880052, | |
| "loss": 0.7254, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.20371577574967406, | |
| "grad_norm": 2.862517833709717, | |
| "learning_rate": 0.0002959256844850065, | |
| "loss": 0.6899, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.24445893089960888, | |
| "grad_norm": 1.9533528089523315, | |
| "learning_rate": 0.0002951108213820078, | |
| "loss": 0.6615, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.28520208604954367, | |
| "grad_norm": 1.9216587543487549, | |
| "learning_rate": 0.0002942959582790091, | |
| "loss": 0.6518, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.3259452411994785, | |
| "grad_norm": 2.019871711730957, | |
| "learning_rate": 0.0002934810951760104, | |
| "loss": 0.6282, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.3666883963494133, | |
| "grad_norm": 2.8539257049560547, | |
| "learning_rate": 0.0002926662320730117, | |
| "loss": 0.6356, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.4074315514993481, | |
| "grad_norm": 2.5572264194488525, | |
| "learning_rate": 0.000291851368970013, | |
| "loss": 0.6119, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.44817470664928294, | |
| "grad_norm": 2.307138681411743, | |
| "learning_rate": 0.0002910365058670143, | |
| "loss": 0.6074, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.48891786179921776, | |
| "grad_norm": 1.5693212747573853, | |
| "learning_rate": 0.0002902216427640156, | |
| "loss": 0.5995, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.5296610169491526, | |
| "grad_norm": 1.5355511903762817, | |
| "learning_rate": 0.00028940677966101693, | |
| "loss": 0.5984, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.5704041720990873, | |
| "grad_norm": 2.0896735191345215, | |
| "learning_rate": 0.0002885919165580182, | |
| "loss": 0.5802, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.6111473272490222, | |
| "grad_norm": 2.17629075050354, | |
| "learning_rate": 0.00028777705345501956, | |
| "loss": 0.5848, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.651890482398957, | |
| "grad_norm": 1.2360671758651733, | |
| "learning_rate": 0.00028696219035202086, | |
| "loss": 0.5867, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.6926336375488917, | |
| "grad_norm": 1.9337974786758423, | |
| "learning_rate": 0.00028614732724902215, | |
| "loss": 0.5769, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.7333767926988266, | |
| "grad_norm": 1.8343297243118286, | |
| "learning_rate": 0.00028533246414602344, | |
| "loss": 0.5785, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.7741199478487614, | |
| "grad_norm": 3.1703989505767822, | |
| "learning_rate": 0.00028451760104302473, | |
| "loss": 0.5726, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.8148631029986962, | |
| "grad_norm": 2.3858680725097656, | |
| "learning_rate": 0.000283702737940026, | |
| "loss": 0.562, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.855606258148631, | |
| "grad_norm": 2.681757688522339, | |
| "learning_rate": 0.00028288787483702737, | |
| "loss": 0.5628, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.8963494132985659, | |
| "grad_norm": 1.7328672409057617, | |
| "learning_rate": 0.00028207301173402866, | |
| "loss": 0.5584, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.9370925684485006, | |
| "grad_norm": 2.7428441047668457, | |
| "learning_rate": 0.00028125814863102995, | |
| "loss": 0.5577, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.9778357235984355, | |
| "grad_norm": 2.219202995300293, | |
| "learning_rate": 0.00028044328552803124, | |
| "loss": 0.554, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7979919910430908, | |
| "eval_loss": 0.5014437437057495, | |
| "eval_runtime": 3.8836, | |
| "eval_samples_per_second": 641.161, | |
| "eval_steps_per_second": 80.338, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 1.0185788787483703, | |
| "grad_norm": 1.668083667755127, | |
| "learning_rate": 0.00027962842242503253, | |
| "loss": 0.5461, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.0593220338983051, | |
| "grad_norm": 2.8752121925354004, | |
| "learning_rate": 0.0002788135593220339, | |
| "loss": 0.5399, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.1000651890482398, | |
| "grad_norm": 1.8229767084121704, | |
| "learning_rate": 0.00027799869621903517, | |
| "loss": 0.5415, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.1408083441981747, | |
| "grad_norm": 1.681829571723938, | |
| "learning_rate": 0.00027718383311603646, | |
| "loss": 0.535, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.1815514993481095, | |
| "grad_norm": 1.5809299945831299, | |
| "learning_rate": 0.0002763689700130378, | |
| "loss": 0.5485, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.2222946544980444, | |
| "grad_norm": 1.6424498558044434, | |
| "learning_rate": 0.0002755541069100391, | |
| "loss": 0.5353, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.263037809647979, | |
| "grad_norm": 2.1335270404815674, | |
| "learning_rate": 0.0002747392438070404, | |
| "loss": 0.5356, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.303780964797914, | |
| "grad_norm": 3.0127382278442383, | |
| "learning_rate": 0.0002739243807040417, | |
| "loss": 0.53, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.3445241199478488, | |
| "grad_norm": 1.603621006011963, | |
| "learning_rate": 0.000273109517601043, | |
| "loss": 0.5298, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.3852672750977835, | |
| "grad_norm": 3.251509428024292, | |
| "learning_rate": 0.0002722946544980443, | |
| "loss": 0.5326, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.4260104302477183, | |
| "grad_norm": 1.379150390625, | |
| "learning_rate": 0.0002714797913950456, | |
| "loss": 0.5316, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.4667535853976532, | |
| "grad_norm": 1.778817892074585, | |
| "learning_rate": 0.0002706649282920469, | |
| "loss": 0.5233, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.5074967405475879, | |
| "grad_norm": 1.4545488357543945, | |
| "learning_rate": 0.00026985006518904825, | |
| "loss": 0.5292, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.548239895697523, | |
| "grad_norm": 1.7037944793701172, | |
| "learning_rate": 0.00026903520208604954, | |
| "loss": 0.5168, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.5889830508474576, | |
| "grad_norm": 2.0902042388916016, | |
| "learning_rate": 0.00026822033898305083, | |
| "loss": 0.53, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.6297262059973925, | |
| "grad_norm": 1.4677634239196777, | |
| "learning_rate": 0.0002674054758800521, | |
| "loss": 0.5258, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.6704693611473274, | |
| "grad_norm": 1.562299132347107, | |
| "learning_rate": 0.0002665906127770534, | |
| "loss": 0.5232, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.711212516297262, | |
| "grad_norm": 2.906895637512207, | |
| "learning_rate": 0.00026577574967405476, | |
| "loss": 0.5153, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.7519556714471969, | |
| "grad_norm": 3.625960350036621, | |
| "learning_rate": 0.00026496088657105605, | |
| "loss": 0.5137, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.7926988265971318, | |
| "grad_norm": 1.401186466217041, | |
| "learning_rate": 0.00026414602346805735, | |
| "loss": 0.5155, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.8334419817470664, | |
| "grad_norm": 4.507143974304199, | |
| "learning_rate": 0.00026333116036505864, | |
| "loss": 0.5239, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.8741851368970013, | |
| "grad_norm": 2.317826986312866, | |
| "learning_rate": 0.00026251629726205993, | |
| "loss": 0.5059, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.9149282920469362, | |
| "grad_norm": 1.684119701385498, | |
| "learning_rate": 0.0002617014341590612, | |
| "loss": 0.5206, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 1.9556714471968708, | |
| "grad_norm": 3.68534255027771, | |
| "learning_rate": 0.00026088657105606257, | |
| "loss": 0.5088, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.996414602346806, | |
| "grad_norm": 1.978193759918213, | |
| "learning_rate": 0.00026007170795306386, | |
| "loss": 0.5077, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8108433485031128, | |
| "eval_loss": 0.48567625880241394, | |
| "eval_runtime": 3.8076, | |
| "eval_samples_per_second": 653.949, | |
| "eval_steps_per_second": 81.941, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 2.0371577574967406, | |
| "grad_norm": 2.52195143699646, | |
| "learning_rate": 0.0002592568448500652, | |
| "loss": 0.4994, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.077900912646675, | |
| "grad_norm": 2.3119261264801025, | |
| "learning_rate": 0.0002584419817470665, | |
| "loss": 0.4989, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.1186440677966103, | |
| "grad_norm": 1.8698070049285889, | |
| "learning_rate": 0.0002576271186440678, | |
| "loss": 0.505, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.159387222946545, | |
| "grad_norm": 1.762832760810852, | |
| "learning_rate": 0.0002568122555410691, | |
| "loss": 0.5061, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.2001303780964796, | |
| "grad_norm": 1.8201305866241455, | |
| "learning_rate": 0.00025599739243807037, | |
| "loss": 0.5009, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.2408735332464147, | |
| "grad_norm": 2.5493578910827637, | |
| "learning_rate": 0.00025518252933507166, | |
| "loss": 0.4942, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.2816166883963493, | |
| "grad_norm": 1.689488172531128, | |
| "learning_rate": 0.000254367666232073, | |
| "loss": 0.505, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.322359843546284, | |
| "grad_norm": 1.5578385591506958, | |
| "learning_rate": 0.0002535528031290743, | |
| "loss": 0.4827, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.363102998696219, | |
| "grad_norm": 1.9981471300125122, | |
| "learning_rate": 0.0002527379400260756, | |
| "loss": 0.4901, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.4038461538461537, | |
| "grad_norm": 3.95417857170105, | |
| "learning_rate": 0.0002519230769230769, | |
| "loss": 0.4954, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.444589308996089, | |
| "grad_norm": 2.437424898147583, | |
| "learning_rate": 0.0002511082138200782, | |
| "loss": 0.5051, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.4853324641460235, | |
| "grad_norm": 2.264540910720825, | |
| "learning_rate": 0.0002502933507170795, | |
| "loss": 0.4995, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.526075619295958, | |
| "grad_norm": 1.7900969982147217, | |
| "learning_rate": 0.0002494784876140808, | |
| "loss": 0.5043, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.5668187744458932, | |
| "grad_norm": 2.914376735687256, | |
| "learning_rate": 0.0002486636245110821, | |
| "loss": 0.501, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.607561929595828, | |
| "grad_norm": 2.3340747356414795, | |
| "learning_rate": 0.00024784876140808345, | |
| "loss": 0.4856, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.648305084745763, | |
| "grad_norm": 1.6408894062042236, | |
| "learning_rate": 0.00024703389830508474, | |
| "loss": 0.4925, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.6890482398956976, | |
| "grad_norm": 2.726527690887451, | |
| "learning_rate": 0.00024621903520208603, | |
| "loss": 0.4925, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.7297913950456323, | |
| "grad_norm": 1.7461756467819214, | |
| "learning_rate": 0.0002454041720990873, | |
| "loss": 0.4949, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 2.770534550195567, | |
| "grad_norm": 1.021328330039978, | |
| "learning_rate": 0.0002445893089960886, | |
| "loss": 0.5003, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.811277705345502, | |
| "grad_norm": 1.8558237552642822, | |
| "learning_rate": 0.00024377444589308996, | |
| "loss": 0.4969, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 2.8520208604954367, | |
| "grad_norm": 1.888190507888794, | |
| "learning_rate": 0.00024295958279009125, | |
| "loss": 0.4763, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.8927640156453718, | |
| "grad_norm": 1.9714635610580444, | |
| "learning_rate": 0.00024214471968709255, | |
| "loss": 0.4823, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 2.9335071707953064, | |
| "grad_norm": 2.248117446899414, | |
| "learning_rate": 0.00024132985658409386, | |
| "loss": 0.4971, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.974250325945241, | |
| "grad_norm": 2.586214065551758, | |
| "learning_rate": 0.00024051499348109516, | |
| "loss": 0.4793, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7767068147659302, | |
| "eval_loss": 0.5534041523933411, | |
| "eval_runtime": 3.8006, | |
| "eval_samples_per_second": 655.161, | |
| "eval_steps_per_second": 82.092, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 3.014993481095176, | |
| "grad_norm": 1.693083643913269, | |
| "learning_rate": 0.00023970013037809645, | |
| "loss": 0.4938, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.055736636245111, | |
| "grad_norm": 1.6453471183776855, | |
| "learning_rate": 0.00023888526727509777, | |
| "loss": 0.479, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 3.0964797913950455, | |
| "grad_norm": 2.535121202468872, | |
| "learning_rate": 0.00023807040417209906, | |
| "loss": 0.4721, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.1372229465449806, | |
| "grad_norm": 2.6568410396575928, | |
| "learning_rate": 0.00023725554106910038, | |
| "loss": 0.4825, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 3.1779661016949152, | |
| "grad_norm": 1.9132291078567505, | |
| "learning_rate": 0.00023644067796610167, | |
| "loss": 0.4706, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 3.21870925684485, | |
| "grad_norm": 1.7388460636138916, | |
| "learning_rate": 0.00023562581486310296, | |
| "loss": 0.4722, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 3.259452411994785, | |
| "grad_norm": 2.760587692260742, | |
| "learning_rate": 0.0002348109517601043, | |
| "loss": 0.4776, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.3001955671447196, | |
| "grad_norm": 1.4599848985671997, | |
| "learning_rate": 0.0002339960886571056, | |
| "loss": 0.4784, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 3.3409387222946547, | |
| "grad_norm": 1.7198021411895752, | |
| "learning_rate": 0.0002331812255541069, | |
| "loss": 0.4727, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 3.3816818774445894, | |
| "grad_norm": 1.6919358968734741, | |
| "learning_rate": 0.0002323663624511082, | |
| "loss": 0.4851, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 3.422425032594524, | |
| "grad_norm": 2.1125240325927734, | |
| "learning_rate": 0.0002315514993481095, | |
| "loss": 0.4593, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.463168187744459, | |
| "grad_norm": 1.8193351030349731, | |
| "learning_rate": 0.00023073663624511082, | |
| "loss": 0.4753, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 3.5039113428943938, | |
| "grad_norm": 1.8467501401901245, | |
| "learning_rate": 0.0002299217731421121, | |
| "loss": 0.4769, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 3.5446544980443284, | |
| "grad_norm": 2.4582698345184326, | |
| "learning_rate": 0.0002291069100391134, | |
| "loss": 0.4725, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 3.5853976531942635, | |
| "grad_norm": 2.6381258964538574, | |
| "learning_rate": 0.00022829204693611472, | |
| "loss": 0.471, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 3.626140808344198, | |
| "grad_norm": 3.26521635055542, | |
| "learning_rate": 0.000227477183833116, | |
| "loss": 0.47, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 3.666883963494133, | |
| "grad_norm": 3.6461341381073, | |
| "learning_rate": 0.0002266623207301173, | |
| "loss": 0.4836, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 3.707627118644068, | |
| "grad_norm": 4.348172664642334, | |
| "learning_rate": 0.00022584745762711862, | |
| "loss": 0.478, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 3.7483702737940026, | |
| "grad_norm": 2.3575916290283203, | |
| "learning_rate": 0.0002250325945241199, | |
| "loss": 0.468, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 3.7891134289439377, | |
| "grad_norm": 1.4927334785461426, | |
| "learning_rate": 0.0002242177314211212, | |
| "loss": 0.4611, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 3.8298565840938723, | |
| "grad_norm": 2.9326066970825195, | |
| "learning_rate": 0.00022340286831812255, | |
| "loss": 0.476, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 3.870599739243807, | |
| "grad_norm": 1.4343719482421875, | |
| "learning_rate": 0.00022258800521512384, | |
| "loss": 0.4792, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 3.9113428943937416, | |
| "grad_norm": 2.0450692176818848, | |
| "learning_rate": 0.00022177314211212516, | |
| "loss": 0.4822, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 3.9520860495436767, | |
| "grad_norm": 1.492274284362793, | |
| "learning_rate": 0.00022095827900912645, | |
| "loss": 0.4602, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 3.9928292046936114, | |
| "grad_norm": 2.1512324810028076, | |
| "learning_rate": 0.00022014341590612774, | |
| "loss": 0.4791, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8224899768829346, | |
| "eval_loss": 0.4442506432533264, | |
| "eval_runtime": 3.8847, | |
| "eval_samples_per_second": 640.977, | |
| "eval_steps_per_second": 80.315, | |
| "step": 49088 | |
| }, | |
| { | |
| "epoch": 4.0335723598435465, | |
| "grad_norm": 1.4296700954437256, | |
| "learning_rate": 0.00021932855280312906, | |
| "loss": 0.4649, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 4.074315514993481, | |
| "grad_norm": 3.133362293243408, | |
| "learning_rate": 0.00021851368970013035, | |
| "loss": 0.4501, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 4.115058670143416, | |
| "grad_norm": 2.441312551498413, | |
| "learning_rate": 0.00021769882659713165, | |
| "loss": 0.4592, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 4.15580182529335, | |
| "grad_norm": 2.3577542304992676, | |
| "learning_rate": 0.00021688396349413296, | |
| "loss": 0.4532, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 4.196544980443286, | |
| "grad_norm": 2.5034992694854736, | |
| "learning_rate": 0.00021606910039113426, | |
| "loss": 0.4537, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 4.237288135593221, | |
| "grad_norm": 2.869847297668457, | |
| "learning_rate": 0.0002152542372881356, | |
| "loss": 0.4558, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 4.278031290743155, | |
| "grad_norm": 2.8850553035736084, | |
| "learning_rate": 0.0002144393741851369, | |
| "loss": 0.4559, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 4.31877444589309, | |
| "grad_norm": 2.0999245643615723, | |
| "learning_rate": 0.00021362451108213819, | |
| "loss": 0.4651, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 4.3595176010430245, | |
| "grad_norm": 2.392293930053711, | |
| "learning_rate": 0.0002128096479791395, | |
| "loss": 0.4621, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 4.400260756192959, | |
| "grad_norm": 1.9911226034164429, | |
| "learning_rate": 0.0002119947848761408, | |
| "loss": 0.4623, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 4.441003911342895, | |
| "grad_norm": 2.728233575820923, | |
| "learning_rate": 0.0002111799217731421, | |
| "loss": 0.4589, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 4.481747066492829, | |
| "grad_norm": 1.9613022804260254, | |
| "learning_rate": 0.0002103650586701434, | |
| "loss": 0.4588, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 4.522490221642764, | |
| "grad_norm": 2.742570161819458, | |
| "learning_rate": 0.0002095501955671447, | |
| "loss": 0.4541, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 4.563233376792699, | |
| "grad_norm": 1.8285661935806274, | |
| "learning_rate": 0.00020873533246414602, | |
| "loss": 0.4563, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 4.603976531942633, | |
| "grad_norm": 2.059859037399292, | |
| "learning_rate": 0.0002079204693611473, | |
| "loss": 0.459, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 4.644719687092568, | |
| "grad_norm": 2.426276445388794, | |
| "learning_rate": 0.0002071056062581486, | |
| "loss": 0.465, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 4.6854628422425035, | |
| "grad_norm": 2.499319553375244, | |
| "learning_rate": 0.00020629074315514992, | |
| "loss": 0.463, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 4.726205997392438, | |
| "grad_norm": 2.2192931175231934, | |
| "learning_rate": 0.0002054758800521512, | |
| "loss": 0.4556, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 4.766949152542373, | |
| "grad_norm": 3.1158642768859863, | |
| "learning_rate": 0.0002046610169491525, | |
| "loss": 0.457, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 4.8076923076923075, | |
| "grad_norm": 1.7781977653503418, | |
| "learning_rate": 0.00020384615384615385, | |
| "loss": 0.4664, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 4.848435462842242, | |
| "grad_norm": 1.9802038669586182, | |
| "learning_rate": 0.00020303129074315514, | |
| "loss": 0.4565, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 4.889178617992178, | |
| "grad_norm": 1.7128177881240845, | |
| "learning_rate": 0.00020221642764015643, | |
| "loss": 0.4573, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 4.929921773142112, | |
| "grad_norm": 3.5915613174438477, | |
| "learning_rate": 0.00020140156453715775, | |
| "loss": 0.4603, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 4.970664928292047, | |
| "grad_norm": 2.1721646785736084, | |
| "learning_rate": 0.00020058670143415904, | |
| "loss": 0.4541, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8257027864456177, | |
| "eval_loss": 0.44336947798728943, | |
| "eval_runtime": 3.8541, | |
| "eval_samples_per_second": 646.057, | |
| "eval_steps_per_second": 80.952, | |
| "step": 61360 | |
| }, | |
| { | |
| "epoch": 5.011408083441982, | |
| "grad_norm": 4.467598915100098, | |
| "learning_rate": 0.00019977183833116036, | |
| "loss": 0.462, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 5.052151238591916, | |
| "grad_norm": 3.215426445007324, | |
| "learning_rate": 0.00019895697522816165, | |
| "loss": 0.4433, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 5.092894393741851, | |
| "grad_norm": 3.5593578815460205, | |
| "learning_rate": 0.00019814211212516294, | |
| "loss": 0.4412, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 5.1336375488917865, | |
| "grad_norm": 1.5453704595565796, | |
| "learning_rate": 0.00019732724902216426, | |
| "loss": 0.4392, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 5.174380704041721, | |
| "grad_norm": 3.003392219543457, | |
| "learning_rate": 0.00019651238591916555, | |
| "loss": 0.4434, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 5.215123859191656, | |
| "grad_norm": 1.394499659538269, | |
| "learning_rate": 0.00019569752281616684, | |
| "loss": 0.4377, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 5.25586701434159, | |
| "grad_norm": 1.8028594255447388, | |
| "learning_rate": 0.00019488265971316816, | |
| "loss": 0.4457, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 5.296610169491525, | |
| "grad_norm": 2.3185994625091553, | |
| "learning_rate": 0.00019406779661016945, | |
| "loss": 0.4396, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 5.337353324641461, | |
| "grad_norm": 2.245239019393921, | |
| "learning_rate": 0.0001932529335071708, | |
| "loss": 0.4548, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 5.378096479791395, | |
| "grad_norm": 1.6174944639205933, | |
| "learning_rate": 0.0001924380704041721, | |
| "loss": 0.4431, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 5.41883963494133, | |
| "grad_norm": 2.5241715908050537, | |
| "learning_rate": 0.00019162320730117338, | |
| "loss": 0.4382, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 5.459582790091265, | |
| "grad_norm": 3.4499781131744385, | |
| "learning_rate": 0.0001908083441981747, | |
| "loss": 0.4552, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 5.500325945241199, | |
| "grad_norm": 2.127242088317871, | |
| "learning_rate": 0.000189993481095176, | |
| "loss": 0.4511, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 5.541069100391134, | |
| "grad_norm": 2.030122756958008, | |
| "learning_rate": 0.00018917861799217729, | |
| "loss": 0.4553, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 5.581812255541069, | |
| "grad_norm": 2.0190391540527344, | |
| "learning_rate": 0.0001883637548891786, | |
| "loss": 0.4471, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 5.622555410691004, | |
| "grad_norm": 3.642808198928833, | |
| "learning_rate": 0.0001875488917861799, | |
| "loss": 0.4494, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 5.663298565840939, | |
| "grad_norm": 2.013524055480957, | |
| "learning_rate": 0.00018673402868318121, | |
| "loss": 0.4425, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 5.704041720990873, | |
| "grad_norm": 3.961810350418091, | |
| "learning_rate": 0.0001859191655801825, | |
| "loss": 0.4438, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 5.744784876140808, | |
| "grad_norm": 1.9334365129470825, | |
| "learning_rate": 0.0001851043024771838, | |
| "loss": 0.4477, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 5.7855280312907436, | |
| "grad_norm": 2.67224383354187, | |
| "learning_rate": 0.00018428943937418514, | |
| "loss": 0.4522, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 5.826271186440678, | |
| "grad_norm": 2.349132776260376, | |
| "learning_rate": 0.00018347457627118644, | |
| "loss": 0.4467, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 5.867014341590613, | |
| "grad_norm": 2.8674731254577637, | |
| "learning_rate": 0.00018265971316818773, | |
| "loss": 0.4379, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 5.9077574967405475, | |
| "grad_norm": 3.5750834941864014, | |
| "learning_rate": 0.00018184485006518905, | |
| "loss": 0.4445, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 5.948500651890482, | |
| "grad_norm": 2.297048330307007, | |
| "learning_rate": 0.00018102998696219034, | |
| "loss": 0.4415, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 5.989243807040417, | |
| "grad_norm": 2.4230237007141113, | |
| "learning_rate": 0.00018021512385919163, | |
| "loss": 0.4405, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.83253014087677, | |
| "eval_loss": 0.42729276418685913, | |
| "eval_runtime": 3.8615, | |
| "eval_samples_per_second": 644.821, | |
| "eval_steps_per_second": 80.797, | |
| "step": 73632 | |
| }, | |
| { | |
| "epoch": 6.029986962190352, | |
| "grad_norm": 1.7936686277389526, | |
| "learning_rate": 0.00017940026075619295, | |
| "loss": 0.437, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 6.070730117340287, | |
| "grad_norm": 2.2741811275482178, | |
| "learning_rate": 0.00017858539765319424, | |
| "loss": 0.434, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 6.111473272490222, | |
| "grad_norm": 2.475470542907715, | |
| "learning_rate": 0.00017777053455019556, | |
| "loss": 0.4298, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 6.152216427640156, | |
| "grad_norm": 1.6614030599594116, | |
| "learning_rate": 0.00017695567144719685, | |
| "loss": 0.43, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 6.192959582790091, | |
| "grad_norm": 2.708757162094116, | |
| "learning_rate": 0.00017614080834419814, | |
| "loss": 0.4335, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 6.2337027379400265, | |
| "grad_norm": 2.0995919704437256, | |
| "learning_rate": 0.00017532594524119946, | |
| "loss": 0.4378, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 6.274445893089961, | |
| "grad_norm": 2.3114981651306152, | |
| "learning_rate": 0.00017451108213820075, | |
| "loss": 0.445, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 6.315189048239896, | |
| "grad_norm": 2.9290952682495117, | |
| "learning_rate": 0.00017369621903520204, | |
| "loss": 0.4257, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 6.3559322033898304, | |
| "grad_norm": 3.3918747901916504, | |
| "learning_rate": 0.0001728813559322034, | |
| "loss": 0.4375, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 6.396675358539765, | |
| "grad_norm": 2.7095093727111816, | |
| "learning_rate": 0.00017206649282920468, | |
| "loss": 0.4464, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 6.4374185136897, | |
| "grad_norm": 1.904373049736023, | |
| "learning_rate": 0.000171251629726206, | |
| "loss": 0.4328, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 6.478161668839635, | |
| "grad_norm": 2.086244583129883, | |
| "learning_rate": 0.0001704367666232073, | |
| "loss": 0.4379, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 6.51890482398957, | |
| "grad_norm": 2.2188174724578857, | |
| "learning_rate": 0.00016962190352020858, | |
| "loss": 0.432, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 6.559647979139505, | |
| "grad_norm": 2.3787903785705566, | |
| "learning_rate": 0.0001688070404172099, | |
| "loss": 0.4368, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 6.600391134289439, | |
| "grad_norm": 2.9371328353881836, | |
| "learning_rate": 0.0001679921773142112, | |
| "loss": 0.4312, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 6.641134289439374, | |
| "grad_norm": 2.65150785446167, | |
| "learning_rate": 0.00016717731421121248, | |
| "loss": 0.4376, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 6.681877444589309, | |
| "grad_norm": 2.1882104873657227, | |
| "learning_rate": 0.0001663624511082138, | |
| "loss": 0.4365, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 6.722620599739244, | |
| "grad_norm": 2.618929862976074, | |
| "learning_rate": 0.0001655475880052151, | |
| "loss": 0.427, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 6.763363754889179, | |
| "grad_norm": 2.831859588623047, | |
| "learning_rate": 0.00016473272490221644, | |
| "loss": 0.4322, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 6.804106910039113, | |
| "grad_norm": 2.4017553329467773, | |
| "learning_rate": 0.00016391786179921773, | |
| "loss": 0.4425, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 6.844850065189048, | |
| "grad_norm": 2.4107367992401123, | |
| "learning_rate": 0.00016310299869621902, | |
| "loss": 0.4261, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 6.885593220338983, | |
| "grad_norm": 1.5284911394119263, | |
| "learning_rate": 0.00016228813559322034, | |
| "loss": 0.428, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 6.926336375488918, | |
| "grad_norm": 1.875391960144043, | |
| "learning_rate": 0.00016147327249022163, | |
| "loss": 0.4248, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 6.967079530638853, | |
| "grad_norm": 2.9802966117858887, | |
| "learning_rate": 0.00016065840938722293, | |
| "loss": 0.4208, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8313252925872803, | |
| "eval_loss": 0.44049832224845886, | |
| "eval_runtime": 3.8652, | |
| "eval_samples_per_second": 644.21, | |
| "eval_steps_per_second": 80.72, | |
| "step": 85904 | |
| }, | |
| { | |
| "epoch": 7.0078226857887875, | |
| "grad_norm": 4.224608898162842, | |
| "learning_rate": 0.00015984354628422424, | |
| "loss": 0.4366, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 7.048565840938722, | |
| "grad_norm": 2.473148822784424, | |
| "learning_rate": 0.00015902868318122554, | |
| "loss": 0.4106, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 7.089308996088657, | |
| "grad_norm": 2.758864164352417, | |
| "learning_rate": 0.00015821382007822685, | |
| "loss": 0.4249, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 7.130052151238592, | |
| "grad_norm": 2.041701555252075, | |
| "learning_rate": 0.00015739895697522815, | |
| "loss": 0.4187, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 7.170795306388527, | |
| "grad_norm": 2.219381093978882, | |
| "learning_rate": 0.00015658409387222944, | |
| "loss": 0.4239, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 7.211538461538462, | |
| "grad_norm": 2.239011526107788, | |
| "learning_rate": 0.00015576923076923076, | |
| "loss": 0.4256, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 7.252281616688396, | |
| "grad_norm": 1.8178561925888062, | |
| "learning_rate": 0.00015495436766623205, | |
| "loss": 0.4273, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 7.293024771838331, | |
| "grad_norm": 3.339010238647461, | |
| "learning_rate": 0.00015413950456323334, | |
| "loss": 0.4273, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 7.333767926988266, | |
| "grad_norm": 1.5560699701309204, | |
| "learning_rate": 0.00015332464146023469, | |
| "loss": 0.4254, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 7.374511082138201, | |
| "grad_norm": 2.376141309738159, | |
| "learning_rate": 0.00015250977835723598, | |
| "loss": 0.4231, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 7.415254237288136, | |
| "grad_norm": 2.4076344966888428, | |
| "learning_rate": 0.00015169491525423727, | |
| "loss": 0.4126, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 7.4559973924380705, | |
| "grad_norm": 1.661089301109314, | |
| "learning_rate": 0.0001508800521512386, | |
| "loss": 0.4224, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 7.496740547588005, | |
| "grad_norm": 2.04123854637146, | |
| "learning_rate": 0.00015006518904823988, | |
| "loss": 0.4272, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 7.53748370273794, | |
| "grad_norm": 1.8965297937393188, | |
| "learning_rate": 0.00014925032594524117, | |
| "loss": 0.4211, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 7.578226857887875, | |
| "grad_norm": 2.6887612342834473, | |
| "learning_rate": 0.0001484354628422425, | |
| "loss": 0.4249, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 7.61897001303781, | |
| "grad_norm": 2.034926414489746, | |
| "learning_rate": 0.0001476205997392438, | |
| "loss": 0.4289, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 7.659713168187745, | |
| "grad_norm": 3.313100814819336, | |
| "learning_rate": 0.0001468057366362451, | |
| "loss": 0.4127, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 7.700456323337679, | |
| "grad_norm": 2.1167919635772705, | |
| "learning_rate": 0.0001459908735332464, | |
| "loss": 0.4264, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 7.741199478487614, | |
| "grad_norm": 2.0670166015625, | |
| "learning_rate": 0.0001451760104302477, | |
| "loss": 0.4317, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 7.781942633637549, | |
| "grad_norm": 3.6086575984954834, | |
| "learning_rate": 0.00014436114732724903, | |
| "loss": 0.4232, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 7.822685788787483, | |
| "grad_norm": 2.8180601596832275, | |
| "learning_rate": 0.00014354628422425032, | |
| "loss": 0.424, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 7.863428943937419, | |
| "grad_norm": 2.9117753505706787, | |
| "learning_rate": 0.0001427314211212516, | |
| "loss": 0.4225, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 7.904172099087353, | |
| "grad_norm": 2.2281785011291504, | |
| "learning_rate": 0.00014191655801825293, | |
| "loss": 0.4236, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 7.944915254237288, | |
| "grad_norm": 3.034166097640991, | |
| "learning_rate": 0.00014110169491525422, | |
| "loss": 0.4283, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 7.985658409387223, | |
| "grad_norm": 2.297738552093506, | |
| "learning_rate": 0.00014028683181225551, | |
| "loss": 0.4278, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8297188878059387, | |
| "eval_loss": 0.43755677342414856, | |
| "eval_runtime": 3.7829, | |
| "eval_samples_per_second": 658.227, | |
| "eval_steps_per_second": 82.477, | |
| "step": 98176 | |
| }, | |
| { | |
| "epoch": 8.026401564537158, | |
| "grad_norm": 2.4886224269866943, | |
| "learning_rate": 0.00013947196870925683, | |
| "loss": 0.4122, | |
| "step": 98500 | |
| }, | |
| { | |
| "epoch": 8.067144719687093, | |
| "grad_norm": 1.8089336156845093, | |
| "learning_rate": 0.00013865710560625815, | |
| "loss": 0.403, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 8.107887874837028, | |
| "grad_norm": 3.1478216648101807, | |
| "learning_rate": 0.00013784224250325944, | |
| "loss": 0.4078, | |
| "step": 99500 | |
| }, | |
| { | |
| "epoch": 8.148631029986962, | |
| "grad_norm": 3.5064048767089844, | |
| "learning_rate": 0.00013702737940026073, | |
| "loss": 0.4212, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 8.189374185136897, | |
| "grad_norm": 3.1338980197906494, | |
| "learning_rate": 0.00013621251629726205, | |
| "loss": 0.4102, | |
| "step": 100500 | |
| }, | |
| { | |
| "epoch": 8.230117340286832, | |
| "grad_norm": 2.583284616470337, | |
| "learning_rate": 0.00013539765319426334, | |
| "loss": 0.4113, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 8.270860495436766, | |
| "grad_norm": 3.1805083751678467, | |
| "learning_rate": 0.00013458279009126466, | |
| "loss": 0.4144, | |
| "step": 101500 | |
| }, | |
| { | |
| "epoch": 8.3116036505867, | |
| "grad_norm": 2.0652964115142822, | |
| "learning_rate": 0.00013376792698826596, | |
| "loss": 0.4077, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 8.352346805736635, | |
| "grad_norm": 1.8258506059646606, | |
| "learning_rate": 0.00013295306388526727, | |
| "loss": 0.4151, | |
| "step": 102500 | |
| }, | |
| { | |
| "epoch": 8.393089960886572, | |
| "grad_norm": 2.371155261993408, | |
| "learning_rate": 0.00013213820078226857, | |
| "loss": 0.4225, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 8.433833116036507, | |
| "grad_norm": 3.617539405822754, | |
| "learning_rate": 0.00013132333767926986, | |
| "loss": 0.4158, | |
| "step": 103500 | |
| }, | |
| { | |
| "epoch": 8.474576271186441, | |
| "grad_norm": 2.629683017730713, | |
| "learning_rate": 0.00013050847457627118, | |
| "loss": 0.4099, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 8.515319426336376, | |
| "grad_norm": 2.359873056411743, | |
| "learning_rate": 0.00012969361147327247, | |
| "loss": 0.4171, | |
| "step": 104500 | |
| }, | |
| { | |
| "epoch": 8.55606258148631, | |
| "grad_norm": 2.8503479957580566, | |
| "learning_rate": 0.00012887874837027379, | |
| "loss": 0.4194, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 8.596805736636245, | |
| "grad_norm": 2.8921594619750977, | |
| "learning_rate": 0.00012806388526727508, | |
| "loss": 0.4125, | |
| "step": 105500 | |
| }, | |
| { | |
| "epoch": 8.63754889178618, | |
| "grad_norm": 1.8355835676193237, | |
| "learning_rate": 0.0001272490221642764, | |
| "loss": 0.412, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 8.678292046936114, | |
| "grad_norm": 3.0607216358184814, | |
| "learning_rate": 0.0001264341590612777, | |
| "loss": 0.4265, | |
| "step": 106500 | |
| }, | |
| { | |
| "epoch": 8.719035202086049, | |
| "grad_norm": 2.338379144668579, | |
| "learning_rate": 0.000125619295958279, | |
| "loss": 0.4142, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 8.759778357235984, | |
| "grad_norm": 2.316218137741089, | |
| "learning_rate": 0.0001248044328552803, | |
| "loss": 0.4099, | |
| "step": 107500 | |
| }, | |
| { | |
| "epoch": 8.800521512385918, | |
| "grad_norm": 2.4564082622528076, | |
| "learning_rate": 0.0001239895697522816, | |
| "loss": 0.4194, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 8.841264667535853, | |
| "grad_norm": 1.8719547986984253, | |
| "learning_rate": 0.0001231747066492829, | |
| "loss": 0.4132, | |
| "step": 108500 | |
| }, | |
| { | |
| "epoch": 8.88200782268579, | |
| "grad_norm": 3.052569627761841, | |
| "learning_rate": 0.00012235984354628423, | |
| "loss": 0.4194, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 8.922750977835724, | |
| "grad_norm": 2.181389808654785, | |
| "learning_rate": 0.0001215449804432855, | |
| "loss": 0.417, | |
| "step": 109500 | |
| }, | |
| { | |
| "epoch": 8.963494132985659, | |
| "grad_norm": 3.063595771789551, | |
| "learning_rate": 0.00012073011734028682, | |
| "loss": 0.422, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.833734929561615, | |
| "eval_loss": 0.43286681175231934, | |
| "eval_runtime": 3.8739, | |
| "eval_samples_per_second": 642.755, | |
| "eval_steps_per_second": 80.538, | |
| "step": 110448 | |
| }, | |
| { | |
| "epoch": 9.004237288135593, | |
| "grad_norm": 2.0804457664489746, | |
| "learning_rate": 0.00011991525423728813, | |
| "loss": 0.4072, | |
| "step": 110500 | |
| }, | |
| { | |
| "epoch": 9.044980443285528, | |
| "grad_norm": 1.9648699760437012, | |
| "learning_rate": 0.00011910039113428943, | |
| "loss": 0.4031, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 9.085723598435463, | |
| "grad_norm": 3.1396656036376953, | |
| "learning_rate": 0.00011828552803129073, | |
| "loss": 0.3942, | |
| "step": 111500 | |
| }, | |
| { | |
| "epoch": 9.126466753585397, | |
| "grad_norm": 2.4233107566833496, | |
| "learning_rate": 0.00011747066492829203, | |
| "loss": 0.4171, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 9.167209908735332, | |
| "grad_norm": 1.7238380908966064, | |
| "learning_rate": 0.00011665580182529335, | |
| "loss": 0.4056, | |
| "step": 112500 | |
| }, | |
| { | |
| "epoch": 9.207953063885267, | |
| "grad_norm": 2.2021853923797607, | |
| "learning_rate": 0.00011584093872229466, | |
| "loss": 0.4089, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 9.248696219035201, | |
| "grad_norm": 2.9419503211975098, | |
| "learning_rate": 0.00011502607561929595, | |
| "loss": 0.399, | |
| "step": 113500 | |
| }, | |
| { | |
| "epoch": 9.289439374185136, | |
| "grad_norm": 2.092937469482422, | |
| "learning_rate": 0.00011421121251629725, | |
| "loss": 0.4113, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 9.330182529335072, | |
| "grad_norm": 2.0860626697540283, | |
| "learning_rate": 0.00011339634941329856, | |
| "loss": 0.4108, | |
| "step": 114500 | |
| }, | |
| { | |
| "epoch": 9.370925684485007, | |
| "grad_norm": 1.9479416608810425, | |
| "learning_rate": 0.00011258148631029986, | |
| "loss": 0.4068, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 9.411668839634942, | |
| "grad_norm": 2.6916277408599854, | |
| "learning_rate": 0.00011176662320730115, | |
| "loss": 0.4061, | |
| "step": 115500 | |
| }, | |
| { | |
| "epoch": 9.452411994784876, | |
| "grad_norm": 1.601837158203125, | |
| "learning_rate": 0.00011095176010430247, | |
| "loss": 0.4074, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 9.493155149934811, | |
| "grad_norm": 2.331357002258301, | |
| "learning_rate": 0.00011013689700130378, | |
| "loss": 0.4118, | |
| "step": 116500 | |
| }, | |
| { | |
| "epoch": 9.533898305084746, | |
| "grad_norm": 2.559669256210327, | |
| "learning_rate": 0.00010932203389830507, | |
| "loss": 0.4142, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 9.57464146023468, | |
| "grad_norm": 2.3902297019958496, | |
| "learning_rate": 0.00010850717079530637, | |
| "loss": 0.3977, | |
| "step": 117500 | |
| }, | |
| { | |
| "epoch": 9.615384615384615, | |
| "grad_norm": 2.525848388671875, | |
| "learning_rate": 0.00010769230769230768, | |
| "loss": 0.4037, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 9.65612777053455, | |
| "grad_norm": 3.530219316482544, | |
| "learning_rate": 0.00010687744458930898, | |
| "loss": 0.4019, | |
| "step": 118500 | |
| }, | |
| { | |
| "epoch": 9.696870925684484, | |
| "grad_norm": 2.741429090499878, | |
| "learning_rate": 0.00010606258148631028, | |
| "loss": 0.4096, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 9.737614080834419, | |
| "grad_norm": 3.318553924560547, | |
| "learning_rate": 0.0001052477183833116, | |
| "loss": 0.4056, | |
| "step": 119500 | |
| }, | |
| { | |
| "epoch": 9.778357235984355, | |
| "grad_norm": 2.6523921489715576, | |
| "learning_rate": 0.0001044328552803129, | |
| "loss": 0.4007, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 9.81910039113429, | |
| "grad_norm": 3.7088372707366943, | |
| "learning_rate": 0.0001036179921773142, | |
| "loss": 0.4016, | |
| "step": 120500 | |
| }, | |
| { | |
| "epoch": 9.859843546284225, | |
| "grad_norm": 2.5519940853118896, | |
| "learning_rate": 0.0001028031290743155, | |
| "loss": 0.4143, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 9.90058670143416, | |
| "grad_norm": 2.149285316467285, | |
| "learning_rate": 0.0001019882659713168, | |
| "loss": 0.4083, | |
| "step": 121500 | |
| }, | |
| { | |
| "epoch": 9.941329856584094, | |
| "grad_norm": 4.22469425201416, | |
| "learning_rate": 0.00010117340286831812, | |
| "loss": 0.404, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 9.982073011734029, | |
| "grad_norm": 2.2363908290863037, | |
| "learning_rate": 0.00010035853976531943, | |
| "loss": 0.4085, | |
| "step": 122500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8401606678962708, | |
| "eval_loss": 0.42958390712738037, | |
| "eval_runtime": 3.786, | |
| "eval_samples_per_second": 657.679, | |
| "eval_steps_per_second": 82.408, | |
| "step": 122720 | |
| }, | |
| { | |
| "epoch": 10.022816166883963, | |
| "grad_norm": 3.384526491165161, | |
| "learning_rate": 9.954367666232072e-05, | |
| "loss": 0.3999, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 10.063559322033898, | |
| "grad_norm": 3.051342725753784, | |
| "learning_rate": 9.872881355932202e-05, | |
| "loss": 0.3996, | |
| "step": 123500 | |
| }, | |
| { | |
| "epoch": 10.104302477183833, | |
| "grad_norm": 3.707674026489258, | |
| "learning_rate": 9.791395045632333e-05, | |
| "loss": 0.4043, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 10.145045632333767, | |
| "grad_norm": 1.7124032974243164, | |
| "learning_rate": 9.709908735332463e-05, | |
| "loss": 0.3918, | |
| "step": 124500 | |
| }, | |
| { | |
| "epoch": 10.185788787483702, | |
| "grad_norm": 2.3350818157196045, | |
| "learning_rate": 9.628422425032592e-05, | |
| "loss": 0.396, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 10.226531942633638, | |
| "grad_norm": 1.8520114421844482, | |
| "learning_rate": 9.546936114732724e-05, | |
| "loss": 0.3906, | |
| "step": 125500 | |
| }, | |
| { | |
| "epoch": 10.267275097783573, | |
| "grad_norm": 2.7649943828582764, | |
| "learning_rate": 9.465449804432855e-05, | |
| "loss": 0.3902, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 10.308018252933508, | |
| "grad_norm": 3.0913712978363037, | |
| "learning_rate": 9.383963494132985e-05, | |
| "loss": 0.4061, | |
| "step": 126500 | |
| }, | |
| { | |
| "epoch": 10.348761408083442, | |
| "grad_norm": 3.6730563640594482, | |
| "learning_rate": 9.302477183833115e-05, | |
| "loss": 0.4036, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 10.389504563233377, | |
| "grad_norm": 2.8968472480773926, | |
| "learning_rate": 9.220990873533245e-05, | |
| "loss": 0.3948, | |
| "step": 127500 | |
| }, | |
| { | |
| "epoch": 10.430247718383312, | |
| "grad_norm": 1.9545537233352661, | |
| "learning_rate": 9.139504563233377e-05, | |
| "loss": 0.4039, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 10.470990873533246, | |
| "grad_norm": 2.1482009887695312, | |
| "learning_rate": 9.058018252933507e-05, | |
| "loss": 0.4032, | |
| "step": 128500 | |
| }, | |
| { | |
| "epoch": 10.51173402868318, | |
| "grad_norm": 4.479248046875, | |
| "learning_rate": 8.976531942633637e-05, | |
| "loss": 0.3827, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 10.552477183833116, | |
| "grad_norm": 2.6518211364746094, | |
| "learning_rate": 8.895045632333767e-05, | |
| "loss": 0.3963, | |
| "step": 129500 | |
| }, | |
| { | |
| "epoch": 10.59322033898305, | |
| "grad_norm": 2.565751314163208, | |
| "learning_rate": 8.813559322033898e-05, | |
| "loss": 0.4135, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 10.633963494132985, | |
| "grad_norm": 3.31779146194458, | |
| "learning_rate": 8.732073011734028e-05, | |
| "loss": 0.4073, | |
| "step": 130500 | |
| }, | |
| { | |
| "epoch": 10.674706649282921, | |
| "grad_norm": 1.9514780044555664, | |
| "learning_rate": 8.650586701434157e-05, | |
| "loss": 0.401, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 10.715449804432856, | |
| "grad_norm": 4.615423679351807, | |
| "learning_rate": 8.569100391134289e-05, | |
| "loss": 0.3972, | |
| "step": 131500 | |
| }, | |
| { | |
| "epoch": 10.75619295958279, | |
| "grad_norm": 3.4876339435577393, | |
| "learning_rate": 8.48761408083442e-05, | |
| "loss": 0.3958, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 10.796936114732725, | |
| "grad_norm": 1.969255805015564, | |
| "learning_rate": 8.406127770534549e-05, | |
| "loss": 0.3997, | |
| "step": 132500 | |
| }, | |
| { | |
| "epoch": 10.83767926988266, | |
| "grad_norm": 3.2826197147369385, | |
| "learning_rate": 8.32464146023468e-05, | |
| "loss": 0.4014, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 10.878422425032594, | |
| "grad_norm": 2.9294662475585938, | |
| "learning_rate": 8.24315514993481e-05, | |
| "loss": 0.3942, | |
| "step": 133500 | |
| }, | |
| { | |
| "epoch": 10.91916558018253, | |
| "grad_norm": 2.2191972732543945, | |
| "learning_rate": 8.161668839634942e-05, | |
| "loss": 0.3821, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 10.959908735332464, | |
| "grad_norm": 2.2126150131225586, | |
| "learning_rate": 8.080182529335071e-05, | |
| "loss": 0.396, | |
| "step": 134500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8333333134651184, | |
| "eval_loss": 0.4348280727863312, | |
| "eval_runtime": 3.8732, | |
| "eval_samples_per_second": 642.875, | |
| "eval_steps_per_second": 80.553, | |
| "step": 134992 | |
| }, | |
| { | |
| "epoch": 11.000651890482398, | |
| "grad_norm": 1.9272228479385376, | |
| "learning_rate": 7.998696219035201e-05, | |
| "loss": 0.409, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 11.041395045632333, | |
| "grad_norm": 3.325286865234375, | |
| "learning_rate": 7.917209908735332e-05, | |
| "loss": 0.3948, | |
| "step": 135500 | |
| }, | |
| { | |
| "epoch": 11.082138200782268, | |
| "grad_norm": 2.996323585510254, | |
| "learning_rate": 7.835723598435462e-05, | |
| "loss": 0.3884, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 11.122881355932204, | |
| "grad_norm": 2.5405139923095703, | |
| "learning_rate": 7.754237288135592e-05, | |
| "loss": 0.3932, | |
| "step": 136500 | |
| }, | |
| { | |
| "epoch": 11.163624511082139, | |
| "grad_norm": 2.4877593517303467, | |
| "learning_rate": 7.672750977835722e-05, | |
| "loss": 0.3908, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 11.204367666232073, | |
| "grad_norm": 2.917015552520752, | |
| "learning_rate": 7.591264667535854e-05, | |
| "loss": 0.3827, | |
| "step": 137500 | |
| }, | |
| { | |
| "epoch": 11.245110821382008, | |
| "grad_norm": 2.060572624206543, | |
| "learning_rate": 7.509778357235985e-05, | |
| "loss": 0.3938, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 11.285853976531943, | |
| "grad_norm": 3.6868770122528076, | |
| "learning_rate": 7.428292046936114e-05, | |
| "loss": 0.3943, | |
| "step": 138500 | |
| }, | |
| { | |
| "epoch": 11.326597131681877, | |
| "grad_norm": 2.118516683578491, | |
| "learning_rate": 7.346805736636244e-05, | |
| "loss": 0.3871, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 11.367340286831812, | |
| "grad_norm": 2.2013978958129883, | |
| "learning_rate": 7.265319426336375e-05, | |
| "loss": 0.3875, | |
| "step": 139500 | |
| }, | |
| { | |
| "epoch": 11.408083441981747, | |
| "grad_norm": 2.284522533416748, | |
| "learning_rate": 7.183833116036505e-05, | |
| "loss": 0.3937, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 11.448826597131681, | |
| "grad_norm": 1.935478925704956, | |
| "learning_rate": 7.102346805736636e-05, | |
| "loss": 0.3933, | |
| "step": 140500 | |
| }, | |
| { | |
| "epoch": 11.489569752281616, | |
| "grad_norm": 3.882283926010132, | |
| "learning_rate": 7.020860495436766e-05, | |
| "loss": 0.3882, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 11.53031290743155, | |
| "grad_norm": 2.2980778217315674, | |
| "learning_rate": 6.939374185136897e-05, | |
| "loss": 0.3994, | |
| "step": 141500 | |
| }, | |
| { | |
| "epoch": 11.571056062581487, | |
| "grad_norm": 3.7042973041534424, | |
| "learning_rate": 6.857887874837027e-05, | |
| "loss": 0.3894, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 11.611799217731422, | |
| "grad_norm": 2.877511739730835, | |
| "learning_rate": 6.776401564537158e-05, | |
| "loss": 0.4033, | |
| "step": 142500 | |
| }, | |
| { | |
| "epoch": 11.652542372881356, | |
| "grad_norm": 3.1929280757904053, | |
| "learning_rate": 6.694915254237287e-05, | |
| "loss": 0.3913, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 11.693285528031291, | |
| "grad_norm": 2.0072107315063477, | |
| "learning_rate": 6.613428943937419e-05, | |
| "loss": 0.3971, | |
| "step": 143500 | |
| }, | |
| { | |
| "epoch": 11.734028683181226, | |
| "grad_norm": 1.9861186742782593, | |
| "learning_rate": 6.531942633637548e-05, | |
| "loss": 0.4003, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 11.77477183833116, | |
| "grad_norm": 2.227025032043457, | |
| "learning_rate": 6.450456323337679e-05, | |
| "loss": 0.4003, | |
| "step": 144500 | |
| }, | |
| { | |
| "epoch": 11.815514993481095, | |
| "grad_norm": 2.0405077934265137, | |
| "learning_rate": 6.368970013037809e-05, | |
| "loss": 0.3931, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 11.85625814863103, | |
| "grad_norm": 3.3660271167755127, | |
| "learning_rate": 6.28748370273794e-05, | |
| "loss": 0.3934, | |
| "step": 145500 | |
| }, | |
| { | |
| "epoch": 11.897001303780964, | |
| "grad_norm": 2.728158473968506, | |
| "learning_rate": 6.20599739243807e-05, | |
| "loss": 0.3843, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 11.937744458930899, | |
| "grad_norm": 2.6212921142578125, | |
| "learning_rate": 6.124511082138199e-05, | |
| "loss": 0.3852, | |
| "step": 146500 | |
| }, | |
| { | |
| "epoch": 11.978487614080834, | |
| "grad_norm": 2.473024368286133, | |
| "learning_rate": 6.0430247718383304e-05, | |
| "loss": 0.3909, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8357429504394531, | |
| "eval_loss": 0.42502352595329285, | |
| "eval_runtime": 3.8531, | |
| "eval_samples_per_second": 646.227, | |
| "eval_steps_per_second": 80.973, | |
| "step": 147264 | |
| }, | |
| { | |
| "epoch": 12.01923076923077, | |
| "grad_norm": 3.567250967025757, | |
| "learning_rate": 5.961538461538461e-05, | |
| "loss": 0.3983, | |
| "step": 147500 | |
| }, | |
| { | |
| "epoch": 12.059973924380705, | |
| "grad_norm": 1.7462067604064941, | |
| "learning_rate": 5.8800521512385915e-05, | |
| "loss": 0.3794, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 12.10071707953064, | |
| "grad_norm": 2.288787364959717, | |
| "learning_rate": 5.798565840938721e-05, | |
| "loss": 0.3925, | |
| "step": 148500 | |
| }, | |
| { | |
| "epoch": 12.141460234680574, | |
| "grad_norm": 3.1145968437194824, | |
| "learning_rate": 5.7170795306388525e-05, | |
| "loss": 0.3793, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 12.182203389830509, | |
| "grad_norm": 2.16363525390625, | |
| "learning_rate": 5.6355932203389824e-05, | |
| "loss": 0.3804, | |
| "step": 149500 | |
| }, | |
| { | |
| "epoch": 12.222946544980443, | |
| "grad_norm": 2.6342670917510986, | |
| "learning_rate": 5.554106910039113e-05, | |
| "loss": 0.3861, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 12.263689700130378, | |
| "grad_norm": 2.9809041023254395, | |
| "learning_rate": 5.4726205997392434e-05, | |
| "loss": 0.3913, | |
| "step": 150500 | |
| }, | |
| { | |
| "epoch": 12.304432855280313, | |
| "grad_norm": 3.3812155723571777, | |
| "learning_rate": 5.391134289439374e-05, | |
| "loss": 0.3857, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 12.345176010430247, | |
| "grad_norm": 2.890817165374756, | |
| "learning_rate": 5.309647979139504e-05, | |
| "loss": 0.393, | |
| "step": 151500 | |
| }, | |
| { | |
| "epoch": 12.385919165580182, | |
| "grad_norm": 3.3339462280273438, | |
| "learning_rate": 5.228161668839635e-05, | |
| "loss": 0.3873, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 12.426662320730117, | |
| "grad_norm": 2.7341129779815674, | |
| "learning_rate": 5.146675358539765e-05, | |
| "loss": 0.3876, | |
| "step": 152500 | |
| }, | |
| { | |
| "epoch": 12.467405475880053, | |
| "grad_norm": 3.096959114074707, | |
| "learning_rate": 5.065189048239895e-05, | |
| "loss": 0.3908, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 12.508148631029988, | |
| "grad_norm": 1.7687112092971802, | |
| "learning_rate": 4.983702737940025e-05, | |
| "loss": 0.394, | |
| "step": 153500 | |
| }, | |
| { | |
| "epoch": 12.548891786179922, | |
| "grad_norm": 2.028165102005005, | |
| "learning_rate": 4.902216427640156e-05, | |
| "loss": 0.3883, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 12.589634941329857, | |
| "grad_norm": 2.485379934310913, | |
| "learning_rate": 4.820730117340286e-05, | |
| "loss": 0.3855, | |
| "step": 154500 | |
| }, | |
| { | |
| "epoch": 12.630378096479792, | |
| "grad_norm": 1.7456655502319336, | |
| "learning_rate": 4.7392438070404173e-05, | |
| "loss": 0.3968, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 12.671121251629726, | |
| "grad_norm": 2.4976985454559326, | |
| "learning_rate": 4.657757496740547e-05, | |
| "loss": 0.3798, | |
| "step": 155500 | |
| }, | |
| { | |
| "epoch": 12.711864406779661, | |
| "grad_norm": 3.9520442485809326, | |
| "learning_rate": 4.576271186440678e-05, | |
| "loss": 0.3894, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 12.752607561929596, | |
| "grad_norm": 2.648386240005493, | |
| "learning_rate": 4.4947848761408075e-05, | |
| "loss": 0.3851, | |
| "step": 156500 | |
| }, | |
| { | |
| "epoch": 12.79335071707953, | |
| "grad_norm": 2.492152690887451, | |
| "learning_rate": 4.413298565840939e-05, | |
| "loss": 0.3822, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 12.834093872229465, | |
| "grad_norm": 2.9962518215179443, | |
| "learning_rate": 4.3318122555410686e-05, | |
| "loss": 0.3858, | |
| "step": 157500 | |
| }, | |
| { | |
| "epoch": 12.8748370273794, | |
| "grad_norm": 2.332040309906006, | |
| "learning_rate": 4.2503259452412e-05, | |
| "loss": 0.3925, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 12.915580182529336, | |
| "grad_norm": 2.6551926136016846, | |
| "learning_rate": 4.1688396349413296e-05, | |
| "loss": 0.3899, | |
| "step": 158500 | |
| }, | |
| { | |
| "epoch": 12.95632333767927, | |
| "grad_norm": 1.7805209159851074, | |
| "learning_rate": 4.08735332464146e-05, | |
| "loss": 0.3885, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 12.997066492829205, | |
| "grad_norm": 2.9438204765319824, | |
| "learning_rate": 4.00586701434159e-05, | |
| "loss": 0.3956, | |
| "step": 159500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8381525874137878, | |
| "eval_loss": 0.4289119243621826, | |
| "eval_runtime": 3.8072, | |
| "eval_samples_per_second": 654.032, | |
| "eval_steps_per_second": 81.951, | |
| "step": 159536 | |
| }, | |
| { | |
| "epoch": 13.03780964797914, | |
| "grad_norm": 2.8224751949310303, | |
| "learning_rate": 3.924380704041721e-05, | |
| "loss": 0.3818, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 13.078552803129075, | |
| "grad_norm": 2.0187859535217285, | |
| "learning_rate": 3.842894393741851e-05, | |
| "loss": 0.3791, | |
| "step": 160500 | |
| }, | |
| { | |
| "epoch": 13.11929595827901, | |
| "grad_norm": 3.1369576454162598, | |
| "learning_rate": 3.761408083441981e-05, | |
| "loss": 0.3799, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 13.160039113428944, | |
| "grad_norm": 2.373286485671997, | |
| "learning_rate": 3.679921773142112e-05, | |
| "loss": 0.3733, | |
| "step": 161500 | |
| }, | |
| { | |
| "epoch": 13.200782268578878, | |
| "grad_norm": 2.583207130432129, | |
| "learning_rate": 3.5984354628422425e-05, | |
| "loss": 0.3958, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 13.241525423728813, | |
| "grad_norm": 2.5118906497955322, | |
| "learning_rate": 3.5169491525423724e-05, | |
| "loss": 0.3915, | |
| "step": 162500 | |
| }, | |
| { | |
| "epoch": 13.282268578878748, | |
| "grad_norm": 5.202625751495361, | |
| "learning_rate": 3.435462842242503e-05, | |
| "loss": 0.3879, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 13.323011734028682, | |
| "grad_norm": 2.9979419708251953, | |
| "learning_rate": 3.3539765319426334e-05, | |
| "loss": 0.3737, | |
| "step": 163500 | |
| }, | |
| { | |
| "epoch": 13.363754889178619, | |
| "grad_norm": 2.0817720890045166, | |
| "learning_rate": 3.272490221642764e-05, | |
| "loss": 0.3912, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 13.404498044328554, | |
| "grad_norm": 2.691849946975708, | |
| "learning_rate": 3.1910039113428944e-05, | |
| "loss": 0.3689, | |
| "step": 164500 | |
| }, | |
| { | |
| "epoch": 13.445241199478488, | |
| "grad_norm": 2.358008861541748, | |
| "learning_rate": 3.109517601043025e-05, | |
| "loss": 0.3793, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 13.485984354628423, | |
| "grad_norm": 2.514547109603882, | |
| "learning_rate": 3.028031290743155e-05, | |
| "loss": 0.3868, | |
| "step": 165500 | |
| }, | |
| { | |
| "epoch": 13.526727509778357, | |
| "grad_norm": 2.5108165740966797, | |
| "learning_rate": 2.9465449804432853e-05, | |
| "loss": 0.385, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 13.567470664928292, | |
| "grad_norm": 3.075470447540283, | |
| "learning_rate": 2.8650586701434158e-05, | |
| "loss": 0.3774, | |
| "step": 166500 | |
| }, | |
| { | |
| "epoch": 13.608213820078227, | |
| "grad_norm": 4.978045463562012, | |
| "learning_rate": 2.7835723598435463e-05, | |
| "loss": 0.3787, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 13.648956975228161, | |
| "grad_norm": 2.699185609817505, | |
| "learning_rate": 2.7020860495436762e-05, | |
| "loss": 0.3772, | |
| "step": 167500 | |
| }, | |
| { | |
| "epoch": 13.689700130378096, | |
| "grad_norm": 1.7423195838928223, | |
| "learning_rate": 2.6205997392438067e-05, | |
| "loss": 0.3838, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 13.73044328552803, | |
| "grad_norm": 1.603785753250122, | |
| "learning_rate": 2.539113428943937e-05, | |
| "loss": 0.3804, | |
| "step": 168500 | |
| }, | |
| { | |
| "epoch": 13.771186440677965, | |
| "grad_norm": 2.6994235515594482, | |
| "learning_rate": 2.4576271186440674e-05, | |
| "loss": 0.3803, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 13.811929595827902, | |
| "grad_norm": 3.038980484008789, | |
| "learning_rate": 2.376140808344198e-05, | |
| "loss": 0.3823, | |
| "step": 169500 | |
| }, | |
| { | |
| "epoch": 13.852672750977836, | |
| "grad_norm": 3.170668840408325, | |
| "learning_rate": 2.294654498044328e-05, | |
| "loss": 0.3803, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 13.893415906127771, | |
| "grad_norm": 2.6691057682037354, | |
| "learning_rate": 2.2131681877444586e-05, | |
| "loss": 0.3809, | |
| "step": 170500 | |
| }, | |
| { | |
| "epoch": 13.934159061277706, | |
| "grad_norm": 1.789117455482483, | |
| "learning_rate": 2.131681877444589e-05, | |
| "loss": 0.3917, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 13.97490221642764, | |
| "grad_norm": 2.110405445098877, | |
| "learning_rate": 2.0501955671447193e-05, | |
| "loss": 0.3792, | |
| "step": 171500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8365461826324463, | |
| "eval_loss": 0.4235801696777344, | |
| "eval_runtime": 3.8552, | |
| "eval_samples_per_second": 645.875, | |
| "eval_steps_per_second": 80.929, | |
| "step": 171808 | |
| }, | |
| { | |
| "epoch": 14.015645371577575, | |
| "grad_norm": 3.1774420738220215, | |
| "learning_rate": 1.9687092568448498e-05, | |
| "loss": 0.3883, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 14.05638852672751, | |
| "grad_norm": 3.016127109527588, | |
| "learning_rate": 1.8872229465449803e-05, | |
| "loss": 0.3798, | |
| "step": 172500 | |
| }, | |
| { | |
| "epoch": 14.097131681877444, | |
| "grad_norm": 1.6009718179702759, | |
| "learning_rate": 1.8057366362451105e-05, | |
| "loss": 0.3712, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 14.137874837027379, | |
| "grad_norm": 3.2171220779418945, | |
| "learning_rate": 1.724250325945241e-05, | |
| "loss": 0.3752, | |
| "step": 173500 | |
| }, | |
| { | |
| "epoch": 14.178617992177314, | |
| "grad_norm": 2.144103765487671, | |
| "learning_rate": 1.6427640156453715e-05, | |
| "loss": 0.3752, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 14.219361147327248, | |
| "grad_norm": 1.7222505807876587, | |
| "learning_rate": 1.5612777053455017e-05, | |
| "loss": 0.389, | |
| "step": 174500 | |
| }, | |
| { | |
| "epoch": 14.260104302477185, | |
| "grad_norm": 1.7213879823684692, | |
| "learning_rate": 1.4797913950456322e-05, | |
| "loss": 0.3752, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 14.30084745762712, | |
| "grad_norm": 3.261892080307007, | |
| "learning_rate": 1.3983050847457626e-05, | |
| "loss": 0.3813, | |
| "step": 175500 | |
| }, | |
| { | |
| "epoch": 14.341590612777054, | |
| "grad_norm": 3.9659616947174072, | |
| "learning_rate": 1.3168187744458931e-05, | |
| "loss": 0.3846, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 14.382333767926989, | |
| "grad_norm": 2.451526403427124, | |
| "learning_rate": 1.2353324641460234e-05, | |
| "loss": 0.3762, | |
| "step": 176500 | |
| }, | |
| { | |
| "epoch": 14.423076923076923, | |
| "grad_norm": 3.402191638946533, | |
| "learning_rate": 1.1538461538461538e-05, | |
| "loss": 0.3783, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 14.463820078226858, | |
| "grad_norm": 1.925841212272644, | |
| "learning_rate": 1.0723598435462841e-05, | |
| "loss": 0.3673, | |
| "step": 177500 | |
| }, | |
| { | |
| "epoch": 14.504563233376793, | |
| "grad_norm": 2.2188963890075684, | |
| "learning_rate": 9.908735332464146e-06, | |
| "loss": 0.3754, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 14.545306388526727, | |
| "grad_norm": 2.604687213897705, | |
| "learning_rate": 9.093872229465448e-06, | |
| "loss": 0.3789, | |
| "step": 178500 | |
| }, | |
| { | |
| "epoch": 14.586049543676662, | |
| "grad_norm": 3.5460221767425537, | |
| "learning_rate": 8.279009126466753e-06, | |
| "loss": 0.3684, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 14.626792698826597, | |
| "grad_norm": 1.9674248695373535, | |
| "learning_rate": 7.464146023468057e-06, | |
| "loss": 0.3732, | |
| "step": 179500 | |
| }, | |
| { | |
| "epoch": 14.667535853976531, | |
| "grad_norm": 2.8252646923065186, | |
| "learning_rate": 6.649282920469361e-06, | |
| "loss": 0.3735, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 14.708279009126468, | |
| "grad_norm": 3.43896746635437, | |
| "learning_rate": 5.834419817470664e-06, | |
| "loss": 0.383, | |
| "step": 180500 | |
| }, | |
| { | |
| "epoch": 14.749022164276402, | |
| "grad_norm": 2.6553750038146973, | |
| "learning_rate": 5.019556714471968e-06, | |
| "loss": 0.3877, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 14.789765319426337, | |
| "grad_norm": 3.002777576446533, | |
| "learning_rate": 4.2046936114732716e-06, | |
| "loss": 0.3757, | |
| "step": 181500 | |
| }, | |
| { | |
| "epoch": 14.830508474576272, | |
| "grad_norm": 3.0359461307525635, | |
| "learning_rate": 3.389830508474576e-06, | |
| "loss": 0.3903, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 14.871251629726206, | |
| "grad_norm": 2.365903615951538, | |
| "learning_rate": 2.5749674054758798e-06, | |
| "loss": 0.3818, | |
| "step": 182500 | |
| }, | |
| { | |
| "epoch": 14.911994784876141, | |
| "grad_norm": 3.1819570064544678, | |
| "learning_rate": 1.7601043024771837e-06, | |
| "loss": 0.3835, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 14.952737940026076, | |
| "grad_norm": 2.5495858192443848, | |
| "learning_rate": 9.452411994784876e-07, | |
| "loss": 0.3751, | |
| "step": 183500 | |
| }, | |
| { | |
| "epoch": 14.99348109517601, | |
| "grad_norm": 2.4265575408935547, | |
| "learning_rate": 1.303780964797914e-07, | |
| "loss": 0.3759, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8361445665359497, | |
| "eval_loss": 0.4246699810028076, | |
| "eval_runtime": 3.8083, | |
| "eval_samples_per_second": 653.84, | |
| "eval_steps_per_second": 81.927, | |
| "step": 184080 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 184080, | |
| "total_flos": 3.921542539724851e+17, | |
| "train_loss": 0.4436988375695671, | |
| "train_runtime": 13089.0372, | |
| "train_samples_per_second": 450.035, | |
| "train_steps_per_second": 14.064 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 184080, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.921542539724851e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |