| { | |
| "best_global_step": 3474, | |
| "best_metric": 5.55628776550293, | |
| "best_model_checkpoint": "./output/checkpoint-3474", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 3474, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "entropy": 3.606692385673523, | |
| "epoch": 0.028785261945883708, | |
| "grad_norm": 3.2999913692474365, | |
| "learning_rate": 4.9e-07, | |
| "loss": 13.6598, | |
| "mean_token_accuracy": 0.16028020828962325, | |
| "num_tokens": 53993.0, | |
| "step": 50 | |
| }, | |
| { | |
| "entropy": 3.618675880432129, | |
| "epoch": 0.057570523891767415, | |
| "grad_norm": 3.101252555847168, | |
| "learning_rate": 9.9e-07, | |
| "loss": 14.0188, | |
| "mean_token_accuracy": 0.1508466500043869, | |
| "num_tokens": 110134.0, | |
| "step": 100 | |
| }, | |
| { | |
| "entropy": 3.5215235900878907, | |
| "epoch": 0.08635578583765112, | |
| "grad_norm": 3.513662815093994, | |
| "learning_rate": 1.49e-06, | |
| "loss": 12.8555, | |
| "mean_token_accuracy": 0.18527640983462335, | |
| "num_tokens": 160191.0, | |
| "step": 150 | |
| }, | |
| { | |
| "entropy": 3.667909698486328, | |
| "epoch": 0.11514104778353483, | |
| "grad_norm": 4.327610492706299, | |
| "learning_rate": 1.99e-06, | |
| "loss": 13.5394, | |
| "mean_token_accuracy": 0.157139780074358, | |
| "num_tokens": 214993.0, | |
| "step": 200 | |
| }, | |
| { | |
| "entropy": 3.768263258934021, | |
| "epoch": 0.14392630972941853, | |
| "grad_norm": 4.290107250213623, | |
| "learning_rate": 1.988450206246317e-06, | |
| "loss": 12.8912, | |
| "mean_token_accuracy": 0.17374794125556947, | |
| "num_tokens": 268184.0, | |
| "step": 250 | |
| }, | |
| { | |
| "entropy": 3.990619196891785, | |
| "epoch": 0.17271157167530224, | |
| "grad_norm": 4.444278717041016, | |
| "learning_rate": 1.976664702416028e-06, | |
| "loss": 12.455, | |
| "mean_token_accuracy": 0.17780130118131637, | |
| "num_tokens": 319458.0, | |
| "step": 300 | |
| }, | |
| { | |
| "entropy": 4.162646284103394, | |
| "epoch": 0.20149683362118595, | |
| "grad_norm": 5.615262508392334, | |
| "learning_rate": 1.9648791985857395e-06, | |
| "loss": 12.0893, | |
| "mean_token_accuracy": 0.18191319867968558, | |
| "num_tokens": 373337.0, | |
| "step": 350 | |
| }, | |
| { | |
| "entropy": 4.532100868225098, | |
| "epoch": 0.23028209556706966, | |
| "grad_norm": 10.074016571044922, | |
| "learning_rate": 1.9530936947554507e-06, | |
| "loss": 11.9261, | |
| "mean_token_accuracy": 0.169477596282959, | |
| "num_tokens": 427526.0, | |
| "step": 400 | |
| }, | |
| { | |
| "entropy": 4.923871030807495, | |
| "epoch": 0.25906735751295334, | |
| "grad_norm": 16.220163345336914, | |
| "learning_rate": 1.9413081909251622e-06, | |
| "loss": 11.0048, | |
| "mean_token_accuracy": 0.1704501649737358, | |
| "num_tokens": 480528.0, | |
| "step": 450 | |
| }, | |
| { | |
| "entropy": 5.521005854606629, | |
| "epoch": 0.28785261945883706, | |
| "grad_norm": 29.904008865356445, | |
| "learning_rate": 1.9295226870948733e-06, | |
| "loss": 9.6524, | |
| "mean_token_accuracy": 0.16450899541378022, | |
| "num_tokens": 535314.0, | |
| "step": 500 | |
| }, | |
| { | |
| "entropy": 6.092623329162597, | |
| "epoch": 0.31663788140472077, | |
| "grad_norm": 17.821575164794922, | |
| "learning_rate": 1.9177371832645845e-06, | |
| "loss": 8.1054, | |
| "mean_token_accuracy": 0.17205011785030366, | |
| "num_tokens": 588410.0, | |
| "step": 550 | |
| }, | |
| { | |
| "entropy": 6.385262680053711, | |
| "epoch": 0.3454231433506045, | |
| "grad_norm": 5.502202987670898, | |
| "learning_rate": 1.9059516794342958e-06, | |
| "loss": 7.4313, | |
| "mean_token_accuracy": 0.1734227080643177, | |
| "num_tokens": 641736.0, | |
| "step": 600 | |
| }, | |
| { | |
| "entropy": 6.278562617301941, | |
| "epoch": 0.3742084052964882, | |
| "grad_norm": 5.4657697677612305, | |
| "learning_rate": 1.8941661756040071e-06, | |
| "loss": 6.9266, | |
| "mean_token_accuracy": 0.18680249139666558, | |
| "num_tokens": 692200.0, | |
| "step": 650 | |
| }, | |
| { | |
| "entropy": 6.553266277313233, | |
| "epoch": 0.4029936672423719, | |
| "grad_norm": 4.955812931060791, | |
| "learning_rate": 1.8823806717737183e-06, | |
| "loss": 6.9847, | |
| "mean_token_accuracy": 0.16679802387952805, | |
| "num_tokens": 745830.0, | |
| "step": 700 | |
| }, | |
| { | |
| "entropy": 6.470935583114624, | |
| "epoch": 0.4317789291882556, | |
| "grad_norm": 4.198381423950195, | |
| "learning_rate": 1.8705951679434296e-06, | |
| "loss": 6.7277, | |
| "mean_token_accuracy": 0.17847734570503235, | |
| "num_tokens": 798872.0, | |
| "step": 750 | |
| }, | |
| { | |
| "entropy": 6.5620588779449465, | |
| "epoch": 0.4605641911341393, | |
| "grad_norm": 3.1793746948242188, | |
| "learning_rate": 1.8588096641131407e-06, | |
| "loss": 6.7032, | |
| "mean_token_accuracy": 0.17336134731769562, | |
| "num_tokens": 853045.0, | |
| "step": 800 | |
| }, | |
| { | |
| "entropy": 6.532204885482788, | |
| "epoch": 0.48934945308002303, | |
| "grad_norm": 3.824537515640259, | |
| "learning_rate": 1.847024160282852e-06, | |
| "loss": 6.5762, | |
| "mean_token_accuracy": 0.1805124071240425, | |
| "num_tokens": 907679.0, | |
| "step": 850 | |
| }, | |
| { | |
| "entropy": 6.535988225936889, | |
| "epoch": 0.5181347150259067, | |
| "grad_norm": 4.350001811981201, | |
| "learning_rate": 1.8352386564525632e-06, | |
| "loss": 6.505, | |
| "mean_token_accuracy": 0.1842605724930763, | |
| "num_tokens": 964170.0, | |
| "step": 900 | |
| }, | |
| { | |
| "entropy": 6.204533562660218, | |
| "epoch": 0.5469199769717904, | |
| "grad_norm": 2.193660020828247, | |
| "learning_rate": 1.8234531526222745e-06, | |
| "loss": 6.1211, | |
| "mean_token_accuracy": 0.21968430042266845, | |
| "num_tokens": 1015909.0, | |
| "step": 950 | |
| }, | |
| { | |
| "entropy": 6.308737449645996, | |
| "epoch": 0.5757052389176741, | |
| "grad_norm": 2.325622320175171, | |
| "learning_rate": 1.8116676487919857e-06, | |
| "loss": 6.1653, | |
| "mean_token_accuracy": 0.21636426240205764, | |
| "num_tokens": 1068859.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "entropy": 6.332560749053955, | |
| "epoch": 0.6044905008635578, | |
| "grad_norm": 2.0439090728759766, | |
| "learning_rate": 1.799882144961697e-06, | |
| "loss": 6.1559, | |
| "mean_token_accuracy": 0.21859725564718246, | |
| "num_tokens": 1123202.0, | |
| "step": 1050 | |
| }, | |
| { | |
| "entropy": 6.042124252319336, | |
| "epoch": 0.6332757628094415, | |
| "grad_norm": 3.621903657913208, | |
| "learning_rate": 1.7880966411314081e-06, | |
| "loss": 5.8441, | |
| "mean_token_accuracy": 0.24906315237283708, | |
| "num_tokens": 1173403.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "entropy": 5.921343173980713, | |
| "epoch": 0.6620610247553252, | |
| "grad_norm": 5.658033847808838, | |
| "learning_rate": 1.7763111373011195e-06, | |
| "loss": 5.7104, | |
| "mean_token_accuracy": 0.2625067520141602, | |
| "num_tokens": 1225026.0, | |
| "step": 1150 | |
| }, | |
| { | |
| "entropy": 6.093586492538452, | |
| "epoch": 0.690846286701209, | |
| "grad_norm": 2.4292995929718018, | |
| "learning_rate": 1.7645256334708308e-06, | |
| "loss": 5.8658, | |
| "mean_token_accuracy": 0.24842385441064835, | |
| "num_tokens": 1279013.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "entropy": 6.119112596511841, | |
| "epoch": 0.7196315486470927, | |
| "grad_norm": 3.369384288787842, | |
| "learning_rate": 1.752740129640542e-06, | |
| "loss": 5.8784, | |
| "mean_token_accuracy": 0.24857850253582, | |
| "num_tokens": 1332547.0, | |
| "step": 1250 | |
| }, | |
| { | |
| "entropy": 6.025163550376892, | |
| "epoch": 0.7484168105929764, | |
| "grad_norm": 2.5110116004943848, | |
| "learning_rate": 1.7409546258102533e-06, | |
| "loss": 5.7769, | |
| "mean_token_accuracy": 0.25835376888513567, | |
| "num_tokens": 1385192.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "entropy": 5.877259612083435, | |
| "epoch": 0.7772020725388601, | |
| "grad_norm": 2.4179303646087646, | |
| "learning_rate": 1.7291691219799646e-06, | |
| "loss": 5.6284, | |
| "mean_token_accuracy": 0.2756252554059029, | |
| "num_tokens": 1437071.0, | |
| "step": 1350 | |
| }, | |
| { | |
| "entropy": 6.002246947288513, | |
| "epoch": 0.8059873344847438, | |
| "grad_norm": 3.494359016418457, | |
| "learning_rate": 1.717383618149676e-06, | |
| "loss": 5.747, | |
| "mean_token_accuracy": 0.26462210685014725, | |
| "num_tokens": 1490818.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "entropy": 5.991955623626709, | |
| "epoch": 0.8347725964306275, | |
| "grad_norm": 2.340975761413574, | |
| "learning_rate": 1.705598114319387e-06, | |
| "loss": 5.7379, | |
| "mean_token_accuracy": 0.26444981098175047, | |
| "num_tokens": 1544997.0, | |
| "step": 1450 | |
| }, | |
| { | |
| "entropy": 5.91768889427185, | |
| "epoch": 0.8635578583765112, | |
| "grad_norm": 2.2394514083862305, | |
| "learning_rate": 1.6938126104890984e-06, | |
| "loss": 5.6564, | |
| "mean_token_accuracy": 0.2730415526032448, | |
| "num_tokens": 1598302.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "entropy": 5.982716989517212, | |
| "epoch": 0.8923431203223949, | |
| "grad_norm": 1.876839518547058, | |
| "learning_rate": 1.6820271066588098e-06, | |
| "loss": 5.7215, | |
| "mean_token_accuracy": 0.26642445534467696, | |
| "num_tokens": 1655267.0, | |
| "step": 1550 | |
| }, | |
| { | |
| "entropy": 5.820467872619629, | |
| "epoch": 0.9211283822682786, | |
| "grad_norm": 2.219966173171997, | |
| "learning_rate": 1.6702416028285209e-06, | |
| "loss": 5.5555, | |
| "mean_token_accuracy": 0.2856418335437775, | |
| "num_tokens": 1709199.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "entropy": 5.996349005699158, | |
| "epoch": 0.9499136442141624, | |
| "grad_norm": 2.247213840484619, | |
| "learning_rate": 1.6584560989982322e-06, | |
| "loss": 5.7283, | |
| "mean_token_accuracy": 0.2696125540137291, | |
| "num_tokens": 1765443.0, | |
| "step": 1650 | |
| }, | |
| { | |
| "entropy": 5.696683068275451, | |
| "epoch": 0.9786989061600461, | |
| "grad_norm": 2.8499979972839355, | |
| "learning_rate": 1.6466705951679433e-06, | |
| "loss": 5.4335, | |
| "mean_token_accuracy": 0.29918427973985673, | |
| "num_tokens": 1817494.0, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_entropy": 5.993559589034401, | |
| "eval_loss": 5.737204551696777, | |
| "eval_mean_token_accuracy": 0.2618687468739699, | |
| "eval_model_preparation_time": 0.0045, | |
| "eval_num_tokens": 1856362.0, | |
| "eval_runtime": 50.5332, | |
| "eval_samples_per_second": 8.588, | |
| "eval_steps_per_second": 4.294, | |
| "step": 1737 | |
| }, | |
| { | |
| "entropy": 5.746842083930969, | |
| "epoch": 1.0074841681059297, | |
| "grad_norm": 2.33052921295166, | |
| "learning_rate": 1.6348850913376547e-06, | |
| "loss": 5.4796, | |
| "mean_token_accuracy": 0.2966849410533905, | |
| "num_tokens": 1870353.0, | |
| "step": 1750 | |
| }, | |
| { | |
| "entropy": 5.859029049873352, | |
| "epoch": 1.0362694300518134, | |
| "grad_norm": 1.6248886585235596, | |
| "learning_rate": 1.6230995875073658e-06, | |
| "loss": 5.5975, | |
| "mean_token_accuracy": 0.2838129925727844, | |
| "num_tokens": 1926205.0, | |
| "step": 1800 | |
| }, | |
| { | |
| "entropy": 5.731445336341858, | |
| "epoch": 1.065054691997697, | |
| "grad_norm": 1.6941566467285156, | |
| "learning_rate": 1.6113140836770771e-06, | |
| "loss": 5.476, | |
| "mean_token_accuracy": 0.2992346465587616, | |
| "num_tokens": 1979821.0, | |
| "step": 1850 | |
| }, | |
| { | |
| "entropy": 5.6993954515457155, | |
| "epoch": 1.0938399539435808, | |
| "grad_norm": 1.1746597290039062, | |
| "learning_rate": 1.5995285798467883e-06, | |
| "loss": 5.4608, | |
| "mean_token_accuracy": 0.3000726142525673, | |
| "num_tokens": 2034373.0, | |
| "step": 1900 | |
| }, | |
| { | |
| "entropy": 5.668873124122619, | |
| "epoch": 1.1226252158894645, | |
| "grad_norm": 1.728211760520935, | |
| "learning_rate": 1.5877430760164996e-06, | |
| "loss": 5.4347, | |
| "mean_token_accuracy": 0.3033922725915909, | |
| "num_tokens": 2087339.0, | |
| "step": 1950 | |
| }, | |
| { | |
| "entropy": 5.624621086120605, | |
| "epoch": 1.1514104778353482, | |
| "grad_norm": 1.4078539609909058, | |
| "learning_rate": 1.5759575721862107e-06, | |
| "loss": 5.3954, | |
| "mean_token_accuracy": 0.30784171640872954, | |
| "num_tokens": 2139520.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "entropy": 5.7141213130950925, | |
| "epoch": 1.180195739781232, | |
| "grad_norm": 2.186459541320801, | |
| "learning_rate": 1.564172068355922e-06, | |
| "loss": 5.4847, | |
| "mean_token_accuracy": 0.29594049394130706, | |
| "num_tokens": 2193987.0, | |
| "step": 2050 | |
| }, | |
| { | |
| "entropy": 5.632415266036987, | |
| "epoch": 1.2089810017271156, | |
| "grad_norm": 1.3601349592208862, | |
| "learning_rate": 1.5523865645256334e-06, | |
| "loss": 5.4135, | |
| "mean_token_accuracy": 0.30366597563028336, | |
| "num_tokens": 2249616.0, | |
| "step": 2100 | |
| }, | |
| { | |
| "entropy": 5.510904269218445, | |
| "epoch": 1.2377662636729994, | |
| "grad_norm": 2.065760612487793, | |
| "learning_rate": 1.5406010606953445e-06, | |
| "loss": 5.2904, | |
| "mean_token_accuracy": 0.3211754837632179, | |
| "num_tokens": 2300863.0, | |
| "step": 2150 | |
| }, | |
| { | |
| "entropy": 5.703383626937867, | |
| "epoch": 1.266551525618883, | |
| "grad_norm": 1.1172698736190796, | |
| "learning_rate": 1.5288155568650559e-06, | |
| "loss": 5.4802, | |
| "mean_token_accuracy": 0.29713701367378237, | |
| "num_tokens": 2356029.0, | |
| "step": 2200 | |
| }, | |
| { | |
| "entropy": 5.565930342674255, | |
| "epoch": 1.2953367875647668, | |
| "grad_norm": 1.7528513669967651, | |
| "learning_rate": 1.5170300530347672e-06, | |
| "loss": 5.3518, | |
| "mean_token_accuracy": 0.31301232606172563, | |
| "num_tokens": 2408957.0, | |
| "step": 2250 | |
| }, | |
| { | |
| "entropy": 5.496430187225342, | |
| "epoch": 1.3241220495106505, | |
| "grad_norm": 1.892640233039856, | |
| "learning_rate": 1.5052445492044786e-06, | |
| "loss": 5.2967, | |
| "mean_token_accuracy": 0.3181899458169937, | |
| "num_tokens": 2462569.0, | |
| "step": 2300 | |
| }, | |
| { | |
| "entropy": 5.725150098800659, | |
| "epoch": 1.3529073114565342, | |
| "grad_norm": 1.774940848350525, | |
| "learning_rate": 1.4934590453741897e-06, | |
| "loss": 5.5215, | |
| "mean_token_accuracy": 0.29055028676986694, | |
| "num_tokens": 2518544.0, | |
| "step": 2350 | |
| }, | |
| { | |
| "entropy": 5.4884827613830565, | |
| "epoch": 1.381692573402418, | |
| "grad_norm": 2.2167599201202393, | |
| "learning_rate": 1.481673541543901e-06, | |
| "loss": 5.2917, | |
| "mean_token_accuracy": 0.31803421139717103, | |
| "num_tokens": 2570863.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "entropy": 5.697079472541809, | |
| "epoch": 1.4104778353483016, | |
| "grad_norm": 1.6489030122756958, | |
| "learning_rate": 1.4698880377136124e-06, | |
| "loss": 5.4982, | |
| "mean_token_accuracy": 0.2925163987278938, | |
| "num_tokens": 2626998.0, | |
| "step": 2450 | |
| }, | |
| { | |
| "entropy": 5.46209939956665, | |
| "epoch": 1.4392630972941853, | |
| "grad_norm": 1.153914451599121, | |
| "learning_rate": 1.4581025338833235e-06, | |
| "loss": 5.2736, | |
| "mean_token_accuracy": 0.3182168474793434, | |
| "num_tokens": 2681568.0, | |
| "step": 2500 | |
| }, | |
| { | |
| "entropy": 5.4405768728256225, | |
| "epoch": 1.468048359240069, | |
| "grad_norm": 3.6614978313446045, | |
| "learning_rate": 1.4463170300530348e-06, | |
| "loss": 5.2515, | |
| "mean_token_accuracy": 0.3218736210465431, | |
| "num_tokens": 2733587.0, | |
| "step": 2550 | |
| }, | |
| { | |
| "entropy": 5.528175053596496, | |
| "epoch": 1.4968336211859528, | |
| "grad_norm": 1.0849746465682983, | |
| "learning_rate": 1.434531526222746e-06, | |
| "loss": 5.3378, | |
| "mean_token_accuracy": 0.31061659604310987, | |
| "num_tokens": 2787003.0, | |
| "step": 2600 | |
| }, | |
| { | |
| "entropy": 5.46110897064209, | |
| "epoch": 1.5256188831318365, | |
| "grad_norm": 1.8315683603286743, | |
| "learning_rate": 1.4227460223924573e-06, | |
| "loss": 5.2782, | |
| "mean_token_accuracy": 0.31781029611825945, | |
| "num_tokens": 2840263.0, | |
| "step": 2650 | |
| }, | |
| { | |
| "entropy": 5.455560960769653, | |
| "epoch": 1.5544041450777202, | |
| "grad_norm": 1.1859091520309448, | |
| "learning_rate": 1.4109605185621684e-06, | |
| "loss": 5.2735, | |
| "mean_token_accuracy": 0.3194814011454582, | |
| "num_tokens": 2894186.0, | |
| "step": 2700 | |
| }, | |
| { | |
| "entropy": 5.430496115684509, | |
| "epoch": 1.583189407023604, | |
| "grad_norm": 2.3500001430511475, | |
| "learning_rate": 1.3991750147318797e-06, | |
| "loss": 5.2464, | |
| "mean_token_accuracy": 0.32140792965888976, | |
| "num_tokens": 2948171.0, | |
| "step": 2750 | |
| }, | |
| { | |
| "entropy": 5.588023023605347, | |
| "epoch": 1.6119746689694876, | |
| "grad_norm": 1.727825403213501, | |
| "learning_rate": 1.3873895109015909e-06, | |
| "loss": 5.4028, | |
| "mean_token_accuracy": 0.3039530631899834, | |
| "num_tokens": 3002678.0, | |
| "step": 2800 | |
| }, | |
| { | |
| "entropy": 5.410525422096253, | |
| "epoch": 1.6407599309153713, | |
| "grad_norm": 1.3401474952697754, | |
| "learning_rate": 1.3756040070713022e-06, | |
| "loss": 5.2298, | |
| "mean_token_accuracy": 0.324065263569355, | |
| "num_tokens": 3055844.0, | |
| "step": 2850 | |
| }, | |
| { | |
| "entropy": 5.36959942817688, | |
| "epoch": 1.669545192861255, | |
| "grad_norm": 1.1892589330673218, | |
| "learning_rate": 1.3638185032410133e-06, | |
| "loss": 5.1956, | |
| "mean_token_accuracy": 0.32639502108097074, | |
| "num_tokens": 3108636.0, | |
| "step": 2900 | |
| }, | |
| { | |
| "entropy": 5.53826907157898, | |
| "epoch": 1.6983304548071387, | |
| "grad_norm": 1.2652360200881958, | |
| "learning_rate": 1.3520329994107247e-06, | |
| "loss": 5.3583, | |
| "mean_token_accuracy": 0.3074926760792732, | |
| "num_tokens": 3162627.0, | |
| "step": 2950 | |
| }, | |
| { | |
| "entropy": 5.417449145317078, | |
| "epoch": 1.7271157167530224, | |
| "grad_norm": 1.584312915802002, | |
| "learning_rate": 1.340247495580436e-06, | |
| "loss": 5.2388, | |
| "mean_token_accuracy": 0.32019727885723115, | |
| "num_tokens": 3216409.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "entropy": 5.241390740871429, | |
| "epoch": 1.7559009786989062, | |
| "grad_norm": 1.5219439268112183, | |
| "learning_rate": 1.3284619917501471e-06, | |
| "loss": 5.0645, | |
| "mean_token_accuracy": 0.3445430138707161, | |
| "num_tokens": 3266967.0, | |
| "step": 3050 | |
| }, | |
| { | |
| "entropy": 5.405424036979675, | |
| "epoch": 1.7846862406447899, | |
| "grad_norm": 2.1165153980255127, | |
| "learning_rate": 1.3166764879198585e-06, | |
| "loss": 5.232, | |
| "mean_token_accuracy": 0.32085000157356264, | |
| "num_tokens": 3319877.0, | |
| "step": 3100 | |
| }, | |
| { | |
| "entropy": 5.123006024360657, | |
| "epoch": 1.8134715025906736, | |
| "grad_norm": 1.2189785242080688, | |
| "learning_rate": 1.3048909840895698e-06, | |
| "loss": 4.9582, | |
| "mean_token_accuracy": 0.356108532845974, | |
| "num_tokens": 3368569.0, | |
| "step": 3150 | |
| }, | |
| { | |
| "entropy": 5.417610831260681, | |
| "epoch": 1.8422567645365573, | |
| "grad_norm": 1.5157604217529297, | |
| "learning_rate": 1.2931054802592812e-06, | |
| "loss": 5.2454, | |
| "mean_token_accuracy": 0.31976755023002623, | |
| "num_tokens": 3422449.0, | |
| "step": 3200 | |
| }, | |
| { | |
| "entropy": 5.409690895080566, | |
| "epoch": 1.871042026482441, | |
| "grad_norm": 1.3088161945343018, | |
| "learning_rate": 1.2813199764289923e-06, | |
| "loss": 5.2348, | |
| "mean_token_accuracy": 0.32325415283441544, | |
| "num_tokens": 3474399.0, | |
| "step": 3250 | |
| }, | |
| { | |
| "entropy": 5.44662567615509, | |
| "epoch": 1.8998272884283247, | |
| "grad_norm": 2.178372621536255, | |
| "learning_rate": 1.2695344725987036e-06, | |
| "loss": 5.2661, | |
| "mean_token_accuracy": 0.3182847076654434, | |
| "num_tokens": 3527726.0, | |
| "step": 3300 | |
| }, | |
| { | |
| "entropy": 5.512614865303039, | |
| "epoch": 1.9286125503742084, | |
| "grad_norm": 1.3050425052642822, | |
| "learning_rate": 1.2577489687684147e-06, | |
| "loss": 5.3416, | |
| "mean_token_accuracy": 0.3084403133392334, | |
| "num_tokens": 3581980.0, | |
| "step": 3350 | |
| }, | |
| { | |
| "entropy": 5.379772834777832, | |
| "epoch": 1.9573978123200921, | |
| "grad_norm": 1.4584404230117798, | |
| "learning_rate": 1.245963464938126e-06, | |
| "loss": 5.2087, | |
| "mean_token_accuracy": 0.32388432770967485, | |
| "num_tokens": 3635393.0, | |
| "step": 3400 | |
| }, | |
| { | |
| "entropy": 5.483665924072266, | |
| "epoch": 1.9861830742659758, | |
| "grad_norm": 1.2157734632492065, | |
| "learning_rate": 1.2341779611078374e-06, | |
| "loss": 5.3101, | |
| "mean_token_accuracy": 0.3121953472495079, | |
| "num_tokens": 3689894.0, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_entropy": 5.711394641805904, | |
| "eval_loss": 5.55628776550293, | |
| "eval_mean_token_accuracy": 0.2764948787777105, | |
| "eval_model_preparation_time": 0.0045, | |
| "eval_num_tokens": 3712724.0, | |
| "eval_runtime": 50.187, | |
| "eval_samples_per_second": 8.648, | |
| "eval_steps_per_second": 4.324, | |
| "step": 3474 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 8685, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 5.088598592372736e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |