| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.9393694421988683, |
| "eval_steps": 500, |
| "global_step": 1200, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.6773952364921569, |
| "epoch": 0.016168148746968473, |
| "grad_norm": 1.5014575719833374, |
| "learning_rate": 1.1062285714285713e-05, |
| "loss": 0.8615, |
| "mean_token_accuracy": 0.8845478415489196, |
| "num_tokens": 637300.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.8168680906295777, |
| "epoch": 0.03233629749393695, |
| "grad_norm": 0.7244049906730652, |
| "learning_rate": 2.3353714285714284e-05, |
| "loss": 0.186, |
| "mean_token_accuracy": 0.9407223045825959, |
| "num_tokens": 1269586.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.8107704758644104, |
| "epoch": 0.04850444624090541, |
| "grad_norm": 0.5931742787361145, |
| "learning_rate": 3.564514285714286e-05, |
| "loss": 0.1683, |
| "mean_token_accuracy": 0.9468494981527329, |
| "num_tokens": 1904757.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.8603073179721832, |
| "epoch": 0.0646725949878739, |
| "grad_norm": 0.9824456572532654, |
| "learning_rate": 4.301911985099481e-05, |
| "loss": 0.1237, |
| "mean_token_accuracy": 0.952459117770195, |
| "num_tokens": 2539356.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.8146251499652862, |
| "epoch": 0.08084074373484236, |
| "grad_norm": 0.3189217150211334, |
| "learning_rate": 4.300921927758375e-05, |
| "loss": 0.1481, |
| "mean_token_accuracy": 0.951959204673767, |
| "num_tokens": 3172047.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.8657496362924576, |
| "epoch": 0.09700889248181083, |
| "grad_norm": 0.33797141909599304, |
| "learning_rate": 4.298832471857447e-05, |
| "loss": 0.0911, |
| "mean_token_accuracy": 0.9599628835916519, |
| "num_tokens": 3810878.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.8595084100961685, |
| "epoch": 0.11317704122877931, |
| "grad_norm": 0.25154030323028564, |
| "learning_rate": 4.295645042272523e-05, |
| "loss": 0.1948, |
| "mean_token_accuracy": 0.9408956855535507, |
| "num_tokens": 4441175.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.8128769963979721, |
| "epoch": 0.1293451899757478, |
| "grad_norm": 0.25640159845352173, |
| "learning_rate": 4.291361812627543e-05, |
| "loss": 0.1197, |
| "mean_token_accuracy": 0.9514931529760361, |
| "num_tokens": 5071862.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.8326072722673417, |
| "epoch": 0.14551333872271624, |
| "grad_norm": 0.20472562313079834, |
| "learning_rate": 4.285985703812284e-05, |
| "loss": 0.1286, |
| "mean_token_accuracy": 0.9528340250253677, |
| "num_tokens": 5703061.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.8518539875745773, |
| "epoch": 0.16168148746968472, |
| "grad_norm": 0.23205353319644928, |
| "learning_rate": 4.2795203819905004e-05, |
| "loss": 0.1255, |
| "mean_token_accuracy": 0.9548479288816452, |
| "num_tokens": 6340603.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.8549301505088807, |
| "epoch": 0.1778496362166532, |
| "grad_norm": 0.2659873068332672, |
| "learning_rate": 4.271970256099837e-05, |
| "loss": 0.1259, |
| "mean_token_accuracy": 0.953066012263298, |
| "num_tokens": 6974337.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.8317383199930191, |
| "epoch": 0.19401778496362165, |
| "grad_norm": 0.44516628980636597, |
| "learning_rate": 4.263340474845209e-05, |
| "loss": 0.1439, |
| "mean_token_accuracy": 0.9494404047727585, |
| "num_tokens": 7613297.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.8361175179481506, |
| "epoch": 0.21018593371059013, |
| "grad_norm": 0.6500918865203857, |
| "learning_rate": 4.253636923187718e-05, |
| "loss": 0.1342, |
| "mean_token_accuracy": 0.9525173217058182, |
| "num_tokens": 8248606.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.8633032321929932, |
| "epoch": 0.22635408245755861, |
| "grad_norm": 0.18834398686885834, |
| "learning_rate": 4.24286621833148e-05, |
| "loss": 0.1119, |
| "mean_token_accuracy": 0.958444133400917, |
| "num_tokens": 8878665.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.8338625460863114, |
| "epoch": 0.2425222312045271, |
| "grad_norm": 0.13926182687282562, |
| "learning_rate": 4.2310357052111164e-05, |
| "loss": 0.1039, |
| "mean_token_accuracy": 0.957411852478981, |
| "num_tokens": 9511238.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.8270522058010101, |
| "epoch": 0.2586903799514956, |
| "grad_norm": 0.18114715814590454, |
| "learning_rate": 4.218153451482976e-05, |
| "loss": 0.144, |
| "mean_token_accuracy": 0.9530034482479095, |
| "num_tokens": 10143404.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.843826699256897, |
| "epoch": 0.274858528698464, |
| "grad_norm": 0.15516184270381927, |
| "learning_rate": 4.2042282420235034e-05, |
| "loss": 0.1734, |
| "mean_token_accuracy": 0.9450976520776748, |
| "num_tokens": 10778284.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.8460217028856277, |
| "epoch": 0.2910266774454325, |
| "grad_norm": 0.2122449427843094, |
| "learning_rate": 4.18926957293852e-05, |
| "loss": 0.1153, |
| "mean_token_accuracy": 0.9557134211063385, |
| "num_tokens": 11417515.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.8405055642127991, |
| "epoch": 0.30719482619240096, |
| "grad_norm": 0.3736467957496643, |
| "learning_rate": 4.173287645087477e-05, |
| "loss": 0.1077, |
| "mean_token_accuracy": 0.9567035138607025, |
| "num_tokens": 12050989.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.8805753111839294, |
| "epoch": 0.32336297493936944, |
| "grad_norm": 0.3333953022956848, |
| "learning_rate": 4.1562933571271246e-05, |
| "loss": 0.1453, |
| "mean_token_accuracy": 0.9514227479696273, |
| "num_tokens": 12683056.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.8673883885145187, |
| "epoch": 0.3395311236863379, |
| "grad_norm": 0.17194989323616028, |
| "learning_rate": 4.138298298079322e-05, |
| "loss": 0.1043, |
| "mean_token_accuracy": 0.9591319918632507, |
| "num_tokens": 13319872.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.8411018908023834, |
| "epoch": 0.3556992724333064, |
| "grad_norm": 0.2302028387784958, |
| "learning_rate": 4.119314739428057e-05, |
| "loss": 0.1336, |
| "mean_token_accuracy": 0.9508422046899796, |
| "num_tokens": 13950795.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.8470937609672546, |
| "epoch": 0.3718674211802749, |
| "grad_norm": 0.19513072073459625, |
| "learning_rate": 4.099355626751084e-05, |
| "loss": 0.1208, |
| "mean_token_accuracy": 0.9550001442432403, |
| "num_tokens": 14582048.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.8418904453516006, |
| "epoch": 0.3880355699272433, |
| "grad_norm": 0.14174064993858337, |
| "learning_rate": 4.0784345708918586e-05, |
| "loss": 0.1092, |
| "mean_token_accuracy": 0.9563545912504197, |
| "num_tokens": 15221101.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.8454319208860397, |
| "epoch": 0.4042037186742118, |
| "grad_norm": 0.17053835093975067, |
| "learning_rate": 4.0565658386778095e-05, |
| "loss": 0.1211, |
| "mean_token_accuracy": 0.955751970410347, |
| "num_tokens": 15850602.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.8607676595449447, |
| "epoch": 0.42037186742118027, |
| "grad_norm": 0.1538228988647461, |
| "learning_rate": 4.033764343191276e-05, |
| "loss": 0.1056, |
| "mean_token_accuracy": 0.9624619007110595, |
| "num_tokens": 16483369.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.8535934597253799, |
| "epoch": 0.43654001616814875, |
| "grad_norm": 0.15055380761623383, |
| "learning_rate": 4.0100456335997316e-05, |
| "loss": 0.084, |
| "mean_token_accuracy": 0.9642143726348877, |
| "num_tokens": 17117847.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.8338070899248123, |
| "epoch": 0.45270816491511723, |
| "grad_norm": 0.1837768405675888, |
| "learning_rate": 3.985425884552241e-05, |
| "loss": 0.0873, |
| "mean_token_accuracy": 0.9619929045438766, |
| "num_tokens": 17753430.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.8260436862707138, |
| "epoch": 0.4688763136620857, |
| "grad_norm": 0.23624153435230255, |
| "learning_rate": 3.959921885149384e-05, |
| "loss": 0.1268, |
| "mean_token_accuracy": 0.9560808658599853, |
| "num_tokens": 18387206.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.8933364123106002, |
| "epoch": 0.4850444624090542, |
| "grad_norm": 0.17879672348499298, |
| "learning_rate": 3.9335510274941545e-05, |
| "loss": 0.0955, |
| "mean_token_accuracy": 0.9601539671421051, |
| "num_tokens": 19027071.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.9035344928503036, |
| "epoch": 0.5012126111560227, |
| "grad_norm": 0.15353450179100037, |
| "learning_rate": 3.906331294831659e-05, |
| "loss": 0.1067, |
| "mean_token_accuracy": 0.9560070037841797, |
| "num_tokens": 19658468.0, |
| "step": 310 |
| }, |
| { |
| "entropy": 0.8730721861124039, |
| "epoch": 0.5173807599029911, |
| "grad_norm": 0.19843076169490814, |
| "learning_rate": 3.878281249285682e-05, |
| "loss": 0.1438, |
| "mean_token_accuracy": 0.9512244701385498, |
| "num_tokens": 20289193.0, |
| "step": 320 |
| }, |
| { |
| "entropy": 0.8783343881368637, |
| "epoch": 0.5335489086499596, |
| "grad_norm": 0.1338294893503189, |
| "learning_rate": 3.8494200192005064e-05, |
| "loss": 0.0806, |
| "mean_token_accuracy": 0.9635367631912232, |
| "num_tokens": 20926533.0, |
| "step": 330 |
| }, |
| { |
| "entropy": 0.8779289960861206, |
| "epoch": 0.549717057396928, |
| "grad_norm": 0.1764674037694931, |
| "learning_rate": 3.819767286096598e-05, |
| "loss": 0.1104, |
| "mean_token_accuracy": 0.9572186172008514, |
| "num_tokens": 21559750.0, |
| "step": 340 |
| }, |
| { |
| "entropy": 0.837685975432396, |
| "epoch": 0.5658852061438965, |
| "grad_norm": 0.20145727694034576, |
| "learning_rate": 3.7893432712490627e-05, |
| "loss": 0.0893, |
| "mean_token_accuracy": 0.9618171989917755, |
| "num_tokens": 22196151.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.8294027328491211, |
| "epoch": 0.582053354890865, |
| "grad_norm": 0.49309608340263367, |
| "learning_rate": 3.758168721898024e-05, |
| "loss": 0.0871, |
| "mean_token_accuracy": 0.9627891510725022, |
| "num_tokens": 22829694.0, |
| "step": 360 |
| }, |
| { |
| "entropy": 0.8539755493402481, |
| "epoch": 0.5982215036378334, |
| "grad_norm": 0.14053401350975037, |
| "learning_rate": 3.726264897100331e-05, |
| "loss": 0.1218, |
| "mean_token_accuracy": 0.9558191239833832, |
| "num_tokens": 23465668.0, |
| "step": 370 |
| }, |
| { |
| "entropy": 0.869808343052864, |
| "epoch": 0.6143896523848019, |
| "grad_norm": 0.5758092403411865, |
| "learning_rate": 3.693653553232229e-05, |
| "loss": 0.1349, |
| "mean_token_accuracy": 0.9525059849023819, |
| "num_tokens": 24100946.0, |
| "step": 380 |
| }, |
| { |
| "entropy": 0.8545525521039963, |
| "epoch": 0.6305578011317704, |
| "grad_norm": 0.1720757931470871, |
| "learning_rate": 3.660356929152904e-05, |
| "loss": 0.1121, |
| "mean_token_accuracy": 0.9561789363622666, |
| "num_tokens": 24739005.0, |
| "step": 390 |
| }, |
| { |
| "entropy": 0.8498503506183624, |
| "epoch": 0.6467259498787389, |
| "grad_norm": 0.13109032809734344, |
| "learning_rate": 3.626397731038991e-05, |
| "loss": 0.1428, |
| "mean_token_accuracy": 0.9503046572208405, |
| "num_tokens": 25376453.0, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.8461112648248672, |
| "epoch": 0.6628940986257074, |
| "grad_norm": 0.15230858325958252, |
| "learning_rate": 3.5917991169004126e-05, |
| "loss": 0.1009, |
| "mean_token_accuracy": 0.9588418394327164, |
| "num_tokens": 26007981.0, |
| "step": 410 |
| }, |
| { |
| "entropy": 0.873944291472435, |
| "epoch": 0.6790622473726758, |
| "grad_norm": 0.13218910992145538, |
| "learning_rate": 3.5565846807880906e-05, |
| "loss": 0.1162, |
| "mean_token_accuracy": 0.9563382536172866, |
| "num_tokens": 26636260.0, |
| "step": 420 |
| }, |
| { |
| "entropy": 0.8747580498456955, |
| "epoch": 0.6952303961196443, |
| "grad_norm": 0.15079568326473236, |
| "learning_rate": 3.5207784367043e-05, |
| "loss": 0.1662, |
| "mean_token_accuracy": 0.9464658439159394, |
| "num_tokens": 27272963.0, |
| "step": 430 |
| }, |
| { |
| "entropy": 0.8773673146963119, |
| "epoch": 0.7113985448666128, |
| "grad_norm": 0.5315278172492981, |
| "learning_rate": 3.484404802226662e-05, |
| "loss": 0.1098, |
| "mean_token_accuracy": 0.9576691329479218, |
| "num_tokens": 27909305.0, |
| "step": 440 |
| }, |
| { |
| "entropy": 0.8305694669485092, |
| "epoch": 0.7275666936135813, |
| "grad_norm": 0.180936798453331, |
| "learning_rate": 3.4474885818568964e-05, |
| "loss": 0.1097, |
| "mean_token_accuracy": 0.9552874326705932, |
| "num_tokens": 28540685.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.8134979695081711, |
| "epoch": 0.7437348423605498, |
| "grad_norm": 0.12312556803226471, |
| "learning_rate": 3.4100549501057445e-05, |
| "loss": 0.1337, |
| "mean_token_accuracy": 0.9502678036689758, |
| "num_tokens": 29175261.0, |
| "step": 460 |
| }, |
| { |
| "entropy": 0.8490791350603104, |
| "epoch": 0.7599029911075182, |
| "grad_norm": 0.13119207322597504, |
| "learning_rate": 3.372129434325548e-05, |
| "loss": 0.1317, |
| "mean_token_accuracy": 0.952011713385582, |
| "num_tokens": 29809342.0, |
| "step": 470 |
| }, |
| { |
| "entropy": 0.8694223254919052, |
| "epoch": 0.7760711398544866, |
| "grad_norm": 0.12799106538295746, |
| "learning_rate": 3.333737897302224e-05, |
| "loss": 0.1037, |
| "mean_token_accuracy": 0.9582739055156708, |
| "num_tokens": 30442885.0, |
| "step": 480 |
| }, |
| { |
| "entropy": 0.8832469940185547, |
| "epoch": 0.7922392886014551, |
| "grad_norm": 0.14574918150901794, |
| "learning_rate": 3.2949065196184916e-05, |
| "loss": 0.083, |
| "mean_token_accuracy": 0.9631638765335083, |
| "num_tokens": 31080641.0, |
| "step": 490 |
| }, |
| { |
| "entropy": 0.8767068654298782, |
| "epoch": 0.8084074373484236, |
| "grad_norm": 0.15174397826194763, |
| "learning_rate": 3.255661781800383e-05, |
| "loss": 0.0968, |
| "mean_token_accuracy": 0.9600045114755631, |
| "num_tokens": 31717033.0, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.8696157097816467, |
| "epoch": 0.824575586095392, |
| "grad_norm": 0.5725116729736328, |
| "learning_rate": 3.2160304462592125e-05, |
| "loss": 0.1107, |
| "mean_token_accuracy": 0.9578035414218903, |
| "num_tokens": 32352930.0, |
| "step": 510 |
| }, |
| { |
| "entropy": 0.8370975911617279, |
| "epoch": 0.8407437348423605, |
| "grad_norm": 0.14664417505264282, |
| "learning_rate": 3.1760395390413126e-05, |
| "loss": 0.112, |
| "mean_token_accuracy": 0.9576535582542419, |
| "num_tokens": 32988877.0, |
| "step": 520 |
| }, |
| { |
| "entropy": 0.8259440779685974, |
| "epoch": 0.856911883589329, |
| "grad_norm": 0.5169638395309448, |
| "learning_rate": 3.1357163313979966e-05, |
| "loss": 0.1127, |
| "mean_token_accuracy": 0.9568720102310181, |
| "num_tokens": 33621971.0, |
| "step": 530 |
| }, |
| { |
| "entropy": 0.8317058831453323, |
| "epoch": 0.8730800323362975, |
| "grad_norm": 0.177755206823349, |
| "learning_rate": 3.0950883211883e-05, |
| "loss": 0.1107, |
| "mean_token_accuracy": 0.9560508698225021, |
| "num_tokens": 34256352.0, |
| "step": 540 |
| }, |
| { |
| "entropy": 0.8352059632539749, |
| "epoch": 0.889248181083266, |
| "grad_norm": 0.15100137889385223, |
| "learning_rate": 3.054183214127192e-05, |
| "loss": 0.1047, |
| "mean_token_accuracy": 0.9574327141046524, |
| "num_tokens": 34893518.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.836120280623436, |
| "epoch": 0.9054163298302345, |
| "grad_norm": 0.1588786393404007, |
| "learning_rate": 3.0130289048920447e-05, |
| "loss": 0.128, |
| "mean_token_accuracy": 0.9500001698732377, |
| "num_tokens": 35529895.0, |
| "step": 560 |
| }, |
| { |
| "entropy": 0.844342041015625, |
| "epoch": 0.9215844785772029, |
| "grad_norm": 0.12973283231258392, |
| "learning_rate": 2.9716534581002384e-05, |
| "loss": 0.1276, |
| "mean_token_accuracy": 0.9518171727657319, |
| "num_tokens": 36168687.0, |
| "step": 570 |
| }, |
| { |
| "entropy": 0.8538145542144775, |
| "epoch": 0.9377526273241714, |
| "grad_norm": 0.15623527765274048, |
| "learning_rate": 2.93008508917088e-05, |
| "loss": 0.0941, |
| "mean_token_accuracy": 0.9619733393192291, |
| "num_tokens": 36799566.0, |
| "step": 580 |
| }, |
| { |
| "entropy": 0.8576534181833267, |
| "epoch": 0.9539207760711399, |
| "grad_norm": 0.2306746244430542, |
| "learning_rate": 2.8883521450836867e-05, |
| "loss": 0.0954, |
| "mean_token_accuracy": 0.9606269925832749, |
| "num_tokens": 37431752.0, |
| "step": 590 |
| }, |
| { |
| "entropy": 0.8665555149316788, |
| "epoch": 0.9700889248181084, |
| "grad_norm": 0.13469445705413818, |
| "learning_rate": 2.8464830850481487e-05, |
| "loss": 0.0987, |
| "mean_token_accuracy": 0.9606908023357391, |
| "num_tokens": 38065700.0, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.8596831768751144, |
| "epoch": 0.9862570735650767, |
| "grad_norm": 0.12054631859064102, |
| "learning_rate": 2.804506461096161e-05, |
| "loss": 0.1001, |
| "mean_token_accuracy": 0.9582655847072601, |
| "num_tokens": 38697017.0, |
| "step": 610 |
| }, |
| { |
| "entropy": 0.8456679488483229, |
| "epoch": 1.001616814874697, |
| "grad_norm": 0.39974308013916016, |
| "learning_rate": 2.7624508986113607e-05, |
| "loss": 0.0878, |
| "mean_token_accuracy": 0.9611430732827437, |
| "num_tokens": 39300859.0, |
| "step": 620 |
| }, |
| { |
| "entropy": 0.8213167607784271, |
| "epoch": 1.0177849636216654, |
| "grad_norm": 0.4462454617023468, |
| "learning_rate": 2.720345076808435e-05, |
| "loss": 0.0956, |
| "mean_token_accuracy": 0.9598243564367295, |
| "num_tokens": 39933837.0, |
| "step": 630 |
| }, |
| { |
| "entropy": 0.7903138369321823, |
| "epoch": 1.0339531123686339, |
| "grad_norm": 0.14300459623336792, |
| "learning_rate": 2.678217709175723e-05, |
| "loss": 0.0877, |
| "mean_token_accuracy": 0.9632098048925399, |
| "num_tokens": 40562061.0, |
| "step": 640 |
| }, |
| { |
| "entropy": 0.7555049926042556, |
| "epoch": 1.0501212611156023, |
| "grad_norm": 0.17101596295833588, |
| "learning_rate": 2.6360975238944493e-05, |
| "loss": 0.0793, |
| "mean_token_accuracy": 0.9643890768289566, |
| "num_tokens": 41189408.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.7405090600252151, |
| "epoch": 1.0662894098625708, |
| "grad_norm": 0.18363745510578156, |
| "learning_rate": 2.5940132442479315e-05, |
| "loss": 0.0932, |
| "mean_token_accuracy": 0.9594863057136536, |
| "num_tokens": 41820476.0, |
| "step": 660 |
| }, |
| { |
| "entropy": 0.7328479170799256, |
| "epoch": 1.0824575586095393, |
| "grad_norm": 0.7928457856178284, |
| "learning_rate": 2.5519935690341314e-05, |
| "loss": 0.1144, |
| "mean_token_accuracy": 0.954973766207695, |
| "num_tokens": 42458986.0, |
| "step": 670 |
| }, |
| { |
| "entropy": 0.7469552457332611, |
| "epoch": 1.0986257073565078, |
| "grad_norm": 0.18894536793231964, |
| "learning_rate": 2.510067152994903e-05, |
| "loss": 0.0886, |
| "mean_token_accuracy": 0.9634099751710892, |
| "num_tokens": 43089118.0, |
| "step": 680 |
| }, |
| { |
| "entropy": 0.7501187264919281, |
| "epoch": 1.1147938561034763, |
| "grad_norm": 0.13618646562099457, |
| "learning_rate": 2.4682625872752833e-05, |
| "loss": 0.0887, |
| "mean_token_accuracy": 0.9628389656543732, |
| "num_tokens": 43718253.0, |
| "step": 690 |
| }, |
| { |
| "entropy": 0.7545085996389389, |
| "epoch": 1.1309620048504447, |
| "grad_norm": 0.13827002048492432, |
| "learning_rate": 2.4266083799261544e-05, |
| "loss": 0.0892, |
| "mean_token_accuracy": 0.9627442210912704, |
| "num_tokens": 44359510.0, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.7485777050256729, |
| "epoch": 1.1471301535974132, |
| "grad_norm": 0.1142207533121109, |
| "learning_rate": 2.3851329364635662e-05, |
| "loss": 0.1079, |
| "mean_token_accuracy": 0.9571398019790649, |
| "num_tokens": 45000462.0, |
| "step": 710 |
| }, |
| { |
| "entropy": 0.7312653630971908, |
| "epoch": 1.1632983023443815, |
| "grad_norm": 0.13678613305091858, |
| "learning_rate": 2.3438645404979872e-05, |
| "loss": 0.0899, |
| "mean_token_accuracy": 0.9621318280696869, |
| "num_tokens": 45634231.0, |
| "step": 720 |
| }, |
| { |
| "entropy": 0.699233689904213, |
| "epoch": 1.17946645109135, |
| "grad_norm": 0.1518629938364029, |
| "learning_rate": 2.3028313344466785e-05, |
| "loss": 0.0909, |
| "mean_token_accuracy": 0.9617949783802032, |
| "num_tokens": 46267266.0, |
| "step": 730 |
| }, |
| { |
| "entropy": 0.6808125108480454, |
| "epoch": 1.1956345998383184, |
| "grad_norm": 0.1525419056415558, |
| "learning_rate": 2.262061300342364e-05, |
| "loss": 0.0953, |
| "mean_token_accuracy": 0.9612728506326675, |
| "num_tokens": 46902449.0, |
| "step": 740 |
| }, |
| { |
| "entropy": 0.6638721615076065, |
| "epoch": 1.211802748585287, |
| "grad_norm": 0.22801771759986877, |
| "learning_rate": 2.2215822407512635e-05, |
| "loss": 0.0982, |
| "mean_token_accuracy": 0.9603689581155777, |
| "num_tokens": 47539529.0, |
| "step": 750 |
| }, |
| { |
| "entropy": 0.6681429892778397, |
| "epoch": 1.2279708973322554, |
| "grad_norm": 0.21193629503250122, |
| "learning_rate": 2.1814217598135126e-05, |
| "loss": 0.0763, |
| "mean_token_accuracy": 0.9644266366958618, |
| "num_tokens": 48176091.0, |
| "step": 760 |
| }, |
| { |
| "entropy": 0.6868150264024735, |
| "epoch": 1.2441390460792239, |
| "grad_norm": 0.1349545121192932, |
| "learning_rate": 2.1416072444188987e-05, |
| "loss": 0.0901, |
| "mean_token_accuracy": 0.9620761543512344, |
| "num_tokens": 48813506.0, |
| "step": 770 |
| }, |
| { |
| "entropy": 0.7375536412000656, |
| "epoch": 1.2603071948261924, |
| "grad_norm": 0.2284357249736786, |
| "learning_rate": 2.102165845530748e-05, |
| "loss": 0.1143, |
| "mean_token_accuracy": 0.9551832169294358, |
| "num_tokens": 49448478.0, |
| "step": 780 |
| }, |
| { |
| "entropy": 0.7473586022853851, |
| "epoch": 1.2764753435731608, |
| "grad_norm": 0.14399640262126923, |
| "learning_rate": 2.0631244596706953e-05, |
| "loss": 0.1009, |
| "mean_token_accuracy": 0.9595120936632157, |
| "num_tokens": 50078361.0, |
| "step": 790 |
| }, |
| { |
| "entropy": 0.7203246772289276, |
| "epoch": 1.2926434923201293, |
| "grad_norm": 0.18459248542785645, |
| "learning_rate": 2.0245097105769725e-05, |
| "loss": 0.1001, |
| "mean_token_accuracy": 0.9596044003963471, |
| "num_tokens": 50716482.0, |
| "step": 800 |
| }, |
| { |
| "entropy": 0.6946973741054535, |
| "epoch": 1.3088116410670978, |
| "grad_norm": 0.17111091315746307, |
| "learning_rate": 1.9863479310487113e-05, |
| "loss": 0.083, |
| "mean_token_accuracy": 0.96545789539814, |
| "num_tokens": 51348158.0, |
| "step": 810 |
| }, |
| { |
| "entropy": 0.6835971772670746, |
| "epoch": 1.3249797898140663, |
| "grad_norm": 0.7274103164672852, |
| "learning_rate": 1.9486651449886568e-05, |
| "loss": 0.0945, |
| "mean_token_accuracy": 0.9608099848031998, |
| "num_tokens": 51979692.0, |
| "step": 820 |
| }, |
| { |
| "entropy": 0.6854576468467712, |
| "epoch": 1.3411479385610348, |
| "grad_norm": 0.7873964309692383, |
| "learning_rate": 1.911487049656517e-05, |
| "loss": 0.1095, |
| "mean_token_accuracy": 0.9552605956792831, |
| "num_tokens": 52615727.0, |
| "step": 830 |
| }, |
| { |
| "entropy": 0.7046183854341507, |
| "epoch": 1.3573160873080032, |
| "grad_norm": 0.16376009583473206, |
| "learning_rate": 1.874838998145075e-05, |
| "loss": 0.0883, |
| "mean_token_accuracy": 0.9624617099761963, |
| "num_tokens": 53249328.0, |
| "step": 840 |
| }, |
| { |
| "entropy": 0.7164037376642227, |
| "epoch": 1.3734842360549717, |
| "grad_norm": 0.5941961407661438, |
| "learning_rate": 1.838745982090991e-05, |
| "loss": 0.0959, |
| "mean_token_accuracy": 0.9604061782360077, |
| "num_tokens": 53880551.0, |
| "step": 850 |
| }, |
| { |
| "entropy": 0.6940079778432846, |
| "epoch": 1.3896523848019402, |
| "grad_norm": 0.13126255571842194, |
| "learning_rate": 1.803232614632098e-05, |
| "loss": 0.103, |
| "mean_token_accuracy": 0.9574708700180053, |
| "num_tokens": 54521844.0, |
| "step": 860 |
| }, |
| { |
| "entropy": 0.7001440197229385, |
| "epoch": 1.4058205335489087, |
| "grad_norm": 0.14183108508586884, |
| "learning_rate": 1.768323113622811e-05, |
| "loss": 0.0836, |
| "mean_token_accuracy": 0.9636358231306076, |
| "num_tokens": 55160077.0, |
| "step": 870 |
| }, |
| { |
| "entropy": 0.6943766325712204, |
| "epoch": 1.4219886822958772, |
| "grad_norm": 0.13666360080242157, |
| "learning_rate": 1.7340412851190937e-05, |
| "loss": 0.0902, |
| "mean_token_accuracy": 0.9626287043094635, |
| "num_tokens": 55791824.0, |
| "step": 880 |
| }, |
| { |
| "entropy": 0.6825817793607711, |
| "epoch": 1.4381568310428456, |
| "grad_norm": 0.15527373552322388, |
| "learning_rate": 1.7004105071442417e-05, |
| "loss": 0.0963, |
| "mean_token_accuracy": 0.9599285125732422, |
| "num_tokens": 56423855.0, |
| "step": 890 |
| }, |
| { |
| "entropy": 0.6809688478708267, |
| "epoch": 1.4543249797898141, |
| "grad_norm": 0.24463769793510437, |
| "learning_rate": 1.667453713746563e-05, |
| "loss": 0.0932, |
| "mean_token_accuracy": 0.9597282737493515, |
| "num_tokens": 57059542.0, |
| "step": 900 |
| }, |
| { |
| "entropy": 0.694112116098404, |
| "epoch": 1.4704931285367826, |
| "grad_norm": 0.15376579761505127, |
| "learning_rate": 1.6351933793598154e-05, |
| "loss": 0.1359, |
| "mean_token_accuracy": 0.9493547439575195, |
| "num_tokens": 57694840.0, |
| "step": 910 |
| }, |
| { |
| "entropy": 0.7208174526691437, |
| "epoch": 1.486661277283751, |
| "grad_norm": 0.17855307459831238, |
| "learning_rate": 1.6036515034770776e-05, |
| "loss": 0.0811, |
| "mean_token_accuracy": 0.9628295093774796, |
| "num_tokens": 58329208.0, |
| "step": 920 |
| }, |
| { |
| "entropy": 0.7264268428087235, |
| "epoch": 1.5028294260307193, |
| "grad_norm": 0.1413581669330597, |
| "learning_rate": 1.5728495956484915e-05, |
| "loss": 0.0895, |
| "mean_token_accuracy": 0.9619016468524932, |
| "num_tokens": 58961666.0, |
| "step": 930 |
| }, |
| { |
| "entropy": 0.7257553398609161, |
| "epoch": 1.5189975747776878, |
| "grad_norm": 0.3373311758041382, |
| "learning_rate": 1.5428086608131232e-05, |
| "loss": 0.0926, |
| "mean_token_accuracy": 0.9613164037466049, |
| "num_tokens": 59595470.0, |
| "step": 940 |
| }, |
| { |
| "entropy": 0.7276300847530365, |
| "epoch": 1.5351657235246563, |
| "grad_norm": 0.1563713103532791, |
| "learning_rate": 1.5135491849749324e-05, |
| "loss": 0.0799, |
| "mean_token_accuracy": 0.9659652322530746, |
| "num_tokens": 60227575.0, |
| "step": 950 |
| }, |
| { |
| "entropy": 0.7257498234510422, |
| "epoch": 1.5513338722716248, |
| "grad_norm": 0.14745177328586578, |
| "learning_rate": 1.4850911212326185e-05, |
| "loss": 0.0894, |
| "mean_token_accuracy": 0.9617056787014008, |
| "num_tokens": 60861583.0, |
| "step": 960 |
| }, |
| { |
| "entropy": 0.7208718538284302, |
| "epoch": 1.5675020210185933, |
| "grad_norm": 0.13323937356472015, |
| "learning_rate": 1.4574538761728838e-05, |
| "loss": 0.0856, |
| "mean_token_accuracy": 0.9639700353145599, |
| "num_tokens": 61490980.0, |
| "step": 970 |
| }, |
| { |
| "entropy": 0.7161700993776321, |
| "epoch": 1.5836701697655617, |
| "grad_norm": 0.5871157050132751, |
| "learning_rate": 1.4306562966363742e-05, |
| "loss": 0.0892, |
| "mean_token_accuracy": 0.9626906663179398, |
| "num_tokens": 62122933.0, |
| "step": 980 |
| }, |
| { |
| "entropy": 0.7074245065450668, |
| "epoch": 1.5998383185125302, |
| "grad_norm": 0.1451254040002823, |
| "learning_rate": 1.4047166568653348e-05, |
| "loss": 0.1084, |
| "mean_token_accuracy": 0.9581695556640625, |
| "num_tokens": 62757235.0, |
| "step": 990 |
| }, |
| { |
| "entropy": 0.708699181675911, |
| "epoch": 1.6160064672594987, |
| "grad_norm": 0.14215019345283508, |
| "learning_rate": 1.3796526460417413e-05, |
| "loss": 0.0833, |
| "mean_token_accuracy": 0.9636478632688522, |
| "num_tokens": 63389026.0, |
| "step": 1000 |
| }, |
| { |
| "entropy": 0.7094081580638886, |
| "epoch": 1.6321746160064672, |
| "grad_norm": 0.14891685545444489, |
| "learning_rate": 1.3554813562244058e-05, |
| "loss": 0.1136, |
| "mean_token_accuracy": 0.9578704237937927, |
| "num_tokens": 64023896.0, |
| "step": 1010 |
| }, |
| { |
| "entropy": 0.7154507130384445, |
| "epoch": 1.6483427647534357, |
| "grad_norm": 0.153501957654953, |
| "learning_rate": 1.3322192706932775e-05, |
| "loss": 0.0945, |
| "mean_token_accuracy": 0.9607921838760376, |
| "num_tokens": 64656038.0, |
| "step": 1020 |
| }, |
| { |
| "entropy": 0.7135091572999954, |
| "epoch": 1.6645109135004041, |
| "grad_norm": 0.1742275357246399, |
| "learning_rate": 1.3098822527089017e-05, |
| "loss": 0.0777, |
| "mean_token_accuracy": 0.966989666223526, |
| "num_tokens": 65291884.0, |
| "step": 1030 |
| }, |
| { |
| "entropy": 0.7181833446025848, |
| "epoch": 1.6806790622473726, |
| "grad_norm": 0.1389065682888031, |
| "learning_rate": 1.288485534694681e-05, |
| "loss": 0.0921, |
| "mean_token_accuracy": 0.9614289939403534, |
| "num_tokens": 65923404.0, |
| "step": 1040 |
| }, |
| { |
| "entropy": 0.7243649184703826, |
| "epoch": 1.696847210994341, |
| "grad_norm": 0.16483035683631897, |
| "learning_rate": 1.2680437078493369e-05, |
| "loss": 0.1133, |
| "mean_token_accuracy": 0.9589362889528275, |
| "num_tokens": 66562107.0, |
| "step": 1050 |
| }, |
| { |
| "entropy": 0.7150461584329605, |
| "epoch": 1.7130153597413096, |
| "grad_norm": 0.13911336660385132, |
| "learning_rate": 1.248570712196638e-05, |
| "loss": 0.0893, |
| "mean_token_accuracy": 0.9613430619239807, |
| "num_tokens": 67203358.0, |
| "step": 1060 |
| }, |
| { |
| "entropy": 0.7125557601451874, |
| "epoch": 1.729183508488278, |
| "grad_norm": 0.1604709029197693, |
| "learning_rate": 1.2300798270791958e-05, |
| "loss": 0.1014, |
| "mean_token_accuracy": 0.9562771052122117, |
| "num_tokens": 67841193.0, |
| "step": 1070 |
| }, |
| { |
| "entropy": 0.7069766879081726, |
| "epoch": 1.7453516572352465, |
| "grad_norm": 0.5357488989830017, |
| "learning_rate": 1.2125836621027959e-05, |
| "loss": 0.0983, |
| "mean_token_accuracy": 0.958621746301651, |
| "num_tokens": 68480860.0, |
| "step": 1080 |
| }, |
| { |
| "entropy": 0.6922294974327088, |
| "epoch": 1.761519805982215, |
| "grad_norm": 0.15722404420375824, |
| "learning_rate": 1.1960941485374512e-05, |
| "loss": 0.0907, |
| "mean_token_accuracy": 0.9600517988204956, |
| "num_tokens": 69116659.0, |
| "step": 1090 |
| }, |
| { |
| "entropy": 0.6841717809438705, |
| "epoch": 1.7776879547291835, |
| "grad_norm": 0.1459880769252777, |
| "learning_rate": 1.1806225311810355e-05, |
| "loss": 0.0892, |
| "mean_token_accuracy": 0.9633763581514359, |
| "num_tokens": 69756306.0, |
| "step": 1100 |
| }, |
| { |
| "entropy": 0.6851155072450638, |
| "epoch": 1.793856103476152, |
| "grad_norm": 0.1419367492198944, |
| "learning_rate": 1.1661793606910487e-05, |
| "loss": 0.0846, |
| "mean_token_accuracy": 0.9632208168506622, |
| "num_tokens": 70392935.0, |
| "step": 1110 |
| }, |
| { |
| "entropy": 0.6891075581312179, |
| "epoch": 1.8100242522231205, |
| "grad_norm": 0.14313970506191254, |
| "learning_rate": 1.1527744863897385e-05, |
| "loss": 0.0868, |
| "mean_token_accuracy": 0.9634499907493591, |
| "num_tokens": 71029903.0, |
| "step": 1120 |
| }, |
| { |
| "entropy": 0.6887877106666564, |
| "epoch": 1.826192400970089, |
| "grad_norm": 0.13888013362884521, |
| "learning_rate": 1.1404170495474916e-05, |
| "loss": 0.1162, |
| "mean_token_accuracy": 0.9550143569707871, |
| "num_tokens": 71669006.0, |
| "step": 1130 |
| }, |
| { |
| "entropy": 0.6871711373329162, |
| "epoch": 1.8423605497170574, |
| "grad_norm": 0.13568949699401855, |
| "learning_rate": 1.1291154771490695e-05, |
| "loss": 0.0851, |
| "mean_token_accuracy": 0.9637445658445358, |
| "num_tokens": 72304607.0, |
| "step": 1140 |
| }, |
| { |
| "entropy": 0.6872768759727478, |
| "epoch": 1.858528698464026, |
| "grad_norm": 0.14399799704551697, |
| "learning_rate": 1.1188774761469405e-05, |
| "loss": 0.1043, |
| "mean_token_accuracy": 0.9601020872592926, |
| "num_tokens": 72937093.0, |
| "step": 1150 |
| }, |
| { |
| "entropy": 0.6857022762298584, |
| "epoch": 1.8746968472109944, |
| "grad_norm": 0.1428367793560028, |
| "learning_rate": 1.1097100282056299e-05, |
| "loss": 0.1017, |
| "mean_token_accuracy": 0.9590070337057114, |
| "num_tokens": 73571103.0, |
| "step": 1160 |
| }, |
| { |
| "entropy": 0.6974191725254059, |
| "epoch": 1.8908649959579629, |
| "grad_norm": 0.3653204143047333, |
| "learning_rate": 1.1016193849406705e-05, |
| "loss": 0.1127, |
| "mean_token_accuracy": 0.9569035053253174, |
| "num_tokens": 74203810.0, |
| "step": 1170 |
| }, |
| { |
| "entropy": 0.7030192404985428, |
| "epoch": 1.9070331447049313, |
| "grad_norm": 0.15754684805870056, |
| "learning_rate": 1.094611063655397e-05, |
| "loss": 0.0847, |
| "mean_token_accuracy": 0.964092680811882, |
| "num_tokens": 74838760.0, |
| "step": 1180 |
| }, |
| { |
| "entropy": 0.695279723405838, |
| "epoch": 1.9232012934518998, |
| "grad_norm": 0.18823955953121185, |
| "learning_rate": 1.0886898435784974e-05, |
| "loss": 0.0847, |
| "mean_token_accuracy": 0.9636576980352402, |
| "num_tokens": 75474515.0, |
| "step": 1190 |
| }, |
| { |
| "entropy": 0.6898054569959641, |
| "epoch": 1.9393694421988683, |
| "grad_norm": 0.34357205033302307, |
| "learning_rate": 1.0838597626048845e-05, |
| "loss": 0.0823, |
| "mean_token_accuracy": 0.9638740360736847, |
| "num_tokens": 76108447.0, |
| "step": 1200 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 1238, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.771449333170307e+18, |
| "train_batch_size": 6, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
