| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 2075, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.060350030175015085, | |
| "grad_norm": 0.47134971618652344, | |
| "learning_rate": 8.18181818181818e-05, | |
| "loss": 2.0011, | |
| "mean_token_accuracy": 0.59705078125, | |
| "num_tokens": 153961.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.12070006035003017, | |
| "grad_norm": 0.469433456659317, | |
| "learning_rate": 0.00016704545454545452, | |
| "loss": 0.9932, | |
| "mean_token_accuracy": 0.7663843035697937, | |
| "num_tokens": 278730.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18105009052504525, | |
| "grad_norm": 0.3573744595050812, | |
| "learning_rate": 0.0002522727272727273, | |
| "loss": 0.6738, | |
| "mean_token_accuracy": 0.8303389132022858, | |
| "num_tokens": 433974.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.24140012070006034, | |
| "grad_norm": 0.4870102107524872, | |
| "learning_rate": 0.0002999887132933212, | |
| "loss": 0.538, | |
| "mean_token_accuracy": 0.8561568921804428, | |
| "num_tokens": 560553.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.30175015087507545, | |
| "grad_norm": 0.44018852710723877, | |
| "learning_rate": 0.0002998791256978121, | |
| "loss": 0.4078, | |
| "mean_token_accuracy": 0.8877666676044464, | |
| "num_tokens": 716164.0, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.3621001810500905, | |
| "grad_norm": 0.43588945269584656, | |
| "learning_rate": 0.0002996530399366737, | |
| "loss": 0.3434, | |
| "mean_token_accuracy": 0.9038257998228073, | |
| "num_tokens": 843777.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4224502112251056, | |
| "grad_norm": 0.3000963032245636, | |
| "learning_rate": 0.00029931063174202567, | |
| "loss": 0.2533, | |
| "mean_token_accuracy": 0.9296510416269302, | |
| "num_tokens": 1000659.0, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.4828002414001207, | |
| "grad_norm": 0.5495067834854126, | |
| "learning_rate": 0.00029885216726118104, | |
| "loss": 0.2361, | |
| "mean_token_accuracy": 0.9325439488887787, | |
| "num_tokens": 1126758.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5431502715751357, | |
| "grad_norm": 0.2725858688354492, | |
| "learning_rate": 0.00029827800284977474, | |
| "loss": 0.1796, | |
| "mean_token_accuracy": 0.9506399631500244, | |
| "num_tokens": 1282319.0, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.6035003017501509, | |
| "grad_norm": 0.36250340938568115, | |
| "learning_rate": 0.00029758858479477575, | |
| "loss": 0.1546, | |
| "mean_token_accuracy": 0.9553977304697037, | |
| "num_tokens": 1410048.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.663850331925166, | |
| "grad_norm": 0.19987879693508148, | |
| "learning_rate": 0.0002967844489675963, | |
| "loss": 0.1278, | |
| "mean_token_accuracy": 0.9646538734436035, | |
| "num_tokens": 1565741.0, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.724200362100181, | |
| "grad_norm": 0.37567138671875, | |
| "learning_rate": 0.00029586622040756957, | |
| "loss": 0.1253, | |
| "mean_token_accuracy": 0.963447117805481, | |
| "num_tokens": 1691742.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7845503922751962, | |
| "grad_norm": 0.3131314218044281, | |
| "learning_rate": 0.0002948346128361186, | |
| "loss": 0.1226, | |
| "mean_token_accuracy": 0.9655348366498947, | |
| "num_tokens": 1849802.0, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.8449004224502112, | |
| "grad_norm": 0.31324562430381775, | |
| "learning_rate": 0.00029369042810199416, | |
| "loss": 0.0962, | |
| "mean_token_accuracy": 0.9730838489532471, | |
| "num_tokens": 1977116.0, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9052504526252263, | |
| "grad_norm": 0.2842608690261841, | |
| "learning_rate": 0.0002924345555580135, | |
| "loss": 0.0959, | |
| "mean_token_accuracy": 0.9728523832559586, | |
| "num_tokens": 2132067.0, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.9656004828002414, | |
| "grad_norm": 0.3550185263156891, | |
| "learning_rate": 0.000291067971369783, | |
| "loss": 0.0935, | |
| "mean_token_accuracy": 0.9737649637460709, | |
| "num_tokens": 2257450.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.091514453291893, | |
| "eval_mean_token_accuracy": 0.9741935675208633, | |
| "eval_num_tokens": 2337343.0, | |
| "eval_runtime": 33.5995, | |
| "eval_samples_per_second": 10.982, | |
| "eval_steps_per_second": 5.506, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.024140012070006, | |
| "grad_norm": 0.16771960258483887, | |
| "learning_rate": 0.0002895917377569438, | |
| "loss": 0.0908, | |
| "mean_token_accuracy": 0.9737480270493891, | |
| "num_tokens": 2406178.0, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.0844900422450212, | |
| "grad_norm": 0.1925903707742691, | |
| "learning_rate": 0.00028800700216752875, | |
| "loss": 0.058, | |
| "mean_token_accuracy": 0.9819272881746293, | |
| "num_tokens": 2547007.0, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.1448400724200363, | |
| "grad_norm": 0.14942210912704468, | |
| "learning_rate": 0.00028631499638607285, | |
| "loss": 0.0826, | |
| "mean_token_accuracy": 0.9763471013307572, | |
| "num_tokens": 2688498.0, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.2051901025950513, | |
| "grad_norm": 0.15795309841632843, | |
| "learning_rate": 0.0002845170355761712, | |
| "loss": 0.0608, | |
| "mean_token_accuracy": 0.9820247828960419, | |
| "num_tokens": 2829301.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.2655401327700664, | |
| "grad_norm": 0.18073949217796326, | |
| "learning_rate": 0.0002826145172582274, | |
| "loss": 0.0734, | |
| "mean_token_accuracy": 0.9791774296760559, | |
| "num_tokens": 2969723.0, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.3258901629450814, | |
| "grad_norm": 0.1643074005842209, | |
| "learning_rate": 0.00028060892022318764, | |
| "loss": 0.0515, | |
| "mean_token_accuracy": 0.9840993517637253, | |
| "num_tokens": 3110675.0, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.3862401931200965, | |
| "grad_norm": 0.12422581762075424, | |
| "learning_rate": 0.0002785018033831051, | |
| "loss": 0.0723, | |
| "mean_token_accuracy": 0.9796582108736038, | |
| "num_tokens": 3250700.0, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.4465902232951118, | |
| "grad_norm": 0.22021085023880005, | |
| "learning_rate": 0.0002762948045594276, | |
| "loss": 0.0531, | |
| "mean_token_accuracy": 0.9839703214168548, | |
| "num_tokens": 3390089.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.5069402534701268, | |
| "grad_norm": 0.10476569831371307, | |
| "learning_rate": 0.0002739896392099502, | |
| "loss": 0.0663, | |
| "mean_token_accuracy": 0.9813223016262055, | |
| "num_tokens": 3528693.0, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.567290283645142, | |
| "grad_norm": 0.19615675508975983, | |
| "learning_rate": 0.00027158809909542307, | |
| "loss": 0.052, | |
| "mean_token_accuracy": 0.9843996149301529, | |
| "num_tokens": 3668997.0, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.627640313820157, | |
| "grad_norm": 0.12505033612251282, | |
| "learning_rate": 0.00026909205088685, | |
| "loss": 0.069, | |
| "mean_token_accuracy": 0.9804346746206284, | |
| "num_tokens": 3810496.0, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.687990343995172, | |
| "grad_norm": 0.154256209731102, | |
| "learning_rate": 0.0002665034347145612, | |
| "loss": 0.053, | |
| "mean_token_accuracy": 0.9843142431974411, | |
| "num_tokens": 3954411.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.748340374170187, | |
| "grad_norm": 0.1239774152636528, | |
| "learning_rate": 0.000263824262660187, | |
| "loss": 0.0609, | |
| "mean_token_accuracy": 0.9821372818946839, | |
| "num_tokens": 4095246.0, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.8086904043452021, | |
| "grad_norm": 0.16891926527023315, | |
| "learning_rate": 0.0002610566171927056, | |
| "loss": 0.0422, | |
| "mean_token_accuracy": 0.9869382613897324, | |
| "num_tokens": 4236888.0, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.8690404345202172, | |
| "grad_norm": 0.1777702122926712, | |
| "learning_rate": 0.00025820264954977976, | |
| "loss": 0.0578, | |
| "mean_token_accuracy": 0.9832831764221192, | |
| "num_tokens": 4377558.0, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.9293904646952322, | |
| "grad_norm": 0.14417409896850586, | |
| "learning_rate": 0.00025526457806564136, | |
| "loss": 0.0437, | |
| "mean_token_accuracy": 0.9862873548269272, | |
| "num_tokens": 4521824.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.9897404948702473, | |
| "grad_norm": 0.15578912198543549, | |
| "learning_rate": 0.00025224468644682245, | |
| "loss": 0.0498, | |
| "mean_token_accuracy": 0.9850816380977631, | |
| "num_tokens": 4654960.0, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.057027868926525116, | |
| "eval_mean_token_accuracy": 0.9834070872616123, | |
| "eval_num_tokens": 4674686.0, | |
| "eval_runtime": 33.5911, | |
| "eval_samples_per_second": 10.985, | |
| "eval_steps_per_second": 5.507, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.048280024140012, | |
| "grad_norm": 0.12735432386398315, | |
| "learning_rate": 0.00024914532199707444, | |
| "loss": 0.0456, | |
| "mean_token_accuracy": 0.9855861381157157, | |
| "num_tokens": 4802199.0, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.1086300543150274, | |
| "grad_norm": 0.07280462980270386, | |
| "learning_rate": 0.00024596889379285353, | |
| "loss": 0.0342, | |
| "mean_token_accuracy": 0.9890019762516021, | |
| "num_tokens": 4934622.0, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.1689800844900424, | |
| "grad_norm": 0.08683612942695618, | |
| "learning_rate": 0.00024271787081079228, | |
| "loss": 0.0483, | |
| "mean_token_accuracy": 0.9849514049291611, | |
| "num_tokens": 5086902.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.2293301146650575, | |
| "grad_norm": 0.15639296174049377, | |
| "learning_rate": 0.00023939478000861117, | |
| "loss": 0.0346, | |
| "mean_token_accuracy": 0.988804299235344, | |
| "num_tokens": 5220193.0, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.2896801448400725, | |
| "grad_norm": 0.10318291187286377, | |
| "learning_rate": 0.00023600220436096318, | |
| "loss": 0.0458, | |
| "mean_token_accuracy": 0.9856922322511673, | |
| "num_tokens": 5370970.0, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.3500301750150876, | |
| "grad_norm": 0.1170400083065033, | |
| "learning_rate": 0.00023254278085173684, | |
| "loss": 0.0322, | |
| "mean_token_accuracy": 0.9894647383689881, | |
| "num_tokens": 5502750.0, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 2.4103802051901027, | |
| "grad_norm": 0.10694035142660141, | |
| "learning_rate": 0.00022901919842437972, | |
| "loss": 0.044, | |
| "mean_token_accuracy": 0.9859293717145919, | |
| "num_tokens": 5650911.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.4707302353651177, | |
| "grad_norm": 0.11283926665782928, | |
| "learning_rate": 0.00022543419589183397, | |
| "loss": 0.034, | |
| "mean_token_accuracy": 0.9892022889852524, | |
| "num_tokens": 5783109.0, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 2.5310802655401328, | |
| "grad_norm": 0.10492369532585144, | |
| "learning_rate": 0.00022179055980770993, | |
| "loss": 0.0482, | |
| "mean_token_accuracy": 0.9849333500862122, | |
| "num_tokens": 5931888.0, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.591430295715148, | |
| "grad_norm": 0.09699155390262604, | |
| "learning_rate": 0.0002180911223003513, | |
| "loss": 0.031, | |
| "mean_token_accuracy": 0.9900956732034684, | |
| "num_tokens": 6063549.0, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 2.651780325890163, | |
| "grad_norm": 0.0748002678155899, | |
| "learning_rate": 0.00021433875887147627, | |
| "loss": 0.0436, | |
| "mean_token_accuracy": 0.9863469779491425, | |
| "num_tokens": 6211960.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.712130356065178, | |
| "grad_norm": 0.09140191972255707, | |
| "learning_rate": 0.00021053638616110525, | |
| "loss": 0.0308, | |
| "mean_token_accuracy": 0.9897224473953247, | |
| "num_tokens": 6344427.0, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 2.772480386240193, | |
| "grad_norm": 0.16827231645584106, | |
| "learning_rate": 0.00020668695968051274, | |
| "loss": 0.0441, | |
| "mean_token_accuracy": 0.9861899012327194, | |
| "num_tokens": 6494950.0, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.832830416415208, | |
| "grad_norm": 0.07273806631565094, | |
| "learning_rate": 0.00020279347151496482, | |
| "loss": 0.0321, | |
| "mean_token_accuracy": 0.9894575864076615, | |
| "num_tokens": 6627853.0, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 2.8931804465902236, | |
| "grad_norm": 0.07277419418096542, | |
| "learning_rate": 0.00019885894799802922, | |
| "loss": 0.0476, | |
| "mean_token_accuracy": 0.985057960152626, | |
| "num_tokens": 6778129.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.9535304767652386, | |
| "grad_norm": 0.07786601781845093, | |
| "learning_rate": 0.00019488644735926396, | |
| "loss": 0.031, | |
| "mean_token_accuracy": 0.9897067219018936, | |
| "num_tokens": 6909983.0, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.047189101576805115, | |
| "eval_mean_token_accuracy": 0.9865975808452916, | |
| "eval_num_tokens": 7012029.0, | |
| "eval_runtime": 33.3626, | |
| "eval_samples_per_second": 11.06, | |
| "eval_steps_per_second": 5.545, | |
| "step": 1245 | |
| }, | |
| { | |
| "epoch": 3.012070006035003, | |
| "grad_norm": 0.06689652055501938, | |
| "learning_rate": 0.00019087905734711452, | |
| "loss": 0.0387, | |
| "mean_token_accuracy": 0.9876031082930025, | |
| "num_tokens": 7048152.0, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.0724200362100182, | |
| "grad_norm": 0.05977766588330269, | |
| "learning_rate": 0.00018683989282886613, | |
| "loss": 0.0264, | |
| "mean_token_accuracy": 0.9908896738290787, | |
| "num_tokens": 7195308.0, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 3.1327700663850333, | |
| "grad_norm": 0.12265495955944061, | |
| "learning_rate": 0.0001827720933695173, | |
| "loss": 0.0338, | |
| "mean_token_accuracy": 0.9888633280992508, | |
| "num_tokens": 7331006.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.1931200965600484, | |
| "grad_norm": 0.06194588169455528, | |
| "learning_rate": 0.00017867882079145627, | |
| "loss": 0.0265, | |
| "mean_token_accuracy": 0.9908727031946182, | |
| "num_tokens": 7479360.0, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 3.2534701267350634, | |
| "grad_norm": 0.11632513999938965, | |
| "learning_rate": 0.00017456325671683724, | |
| "loss": 0.0324, | |
| "mean_token_accuracy": 0.9889560562372207, | |
| "num_tokens": 7614584.0, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.3138201569100785, | |
| "grad_norm": 0.07189938426017761, | |
| "learning_rate": 0.00017042860009456638, | |
| "loss": 0.0289, | |
| "mean_token_accuracy": 0.9903894019126892, | |
| "num_tokens": 7761823.0, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 3.3741701870850935, | |
| "grad_norm": 0.12062046676874161, | |
| "learning_rate": 0.00016627806471382066, | |
| "loss": 0.0325, | |
| "mean_token_accuracy": 0.989678715467453, | |
| "num_tokens": 7895960.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.4345202172601086, | |
| "grad_norm": 0.09952951222658157, | |
| "learning_rate": 0.00016211487670603078, | |
| "loss": 0.0272, | |
| "mean_token_accuracy": 0.9904452234506607, | |
| "num_tokens": 8039619.0, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 3.4948702474351236, | |
| "grad_norm": 0.09827233850955963, | |
| "learning_rate": 0.0001579422720372715, | |
| "loss": 0.0335, | |
| "mean_token_accuracy": 0.9891135483980179, | |
| "num_tokens": 8173998.0, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 3.5552202776101387, | |
| "grad_norm": 0.06022663414478302, | |
| "learning_rate": 0.00015376349399300745, | |
| "loss": 0.0286, | |
| "mean_token_accuracy": 0.9904164570569992, | |
| "num_tokens": 8320886.0, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 3.6155703077851538, | |
| "grad_norm": 0.16854898631572723, | |
| "learning_rate": 0.0001495817906571492, | |
| "loss": 0.0312, | |
| "mean_token_accuracy": 0.9896230679750443, | |
| "num_tokens": 8457172.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.675920337960169, | |
| "grad_norm": 0.08314011245965958, | |
| "learning_rate": 0.00014540041238738055, | |
| "loss": 0.0286, | |
| "mean_token_accuracy": 0.9901280373334884, | |
| "num_tokens": 8606196.0, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 3.736270368135184, | |
| "grad_norm": 0.0902101993560791, | |
| "learning_rate": 0.00014122260928871734, | |
| "loss": 0.0298, | |
| "mean_token_accuracy": 0.989647666811943, | |
| "num_tokens": 8742275.0, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.796620398310199, | |
| "grad_norm": 0.0578254871070385, | |
| "learning_rate": 0.00013705162868726396, | |
| "loss": 0.0253, | |
| "mean_token_accuracy": 0.9909042817354202, | |
| "num_tokens": 8888173.0, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 3.856970428485214, | |
| "grad_norm": 0.11160247772932053, | |
| "learning_rate": 0.00013289071260612855, | |
| "loss": 0.0289, | |
| "mean_token_accuracy": 0.9900320452451706, | |
| "num_tokens": 9023700.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.9173204586602295, | |
| "grad_norm": 0.05397836118936539, | |
| "learning_rate": 0.00012874309524546083, | |
| "loss": 0.0258, | |
| "mean_token_accuracy": 0.9909697133302688, | |
| "num_tokens": 9171113.0, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 3.9776704888352445, | |
| "grad_norm": 0.058956462889909744, | |
| "learning_rate": 0.00012461200046857084, | |
| "loss": 0.0254, | |
| "mean_token_accuracy": 0.9911587375402451, | |
| "num_tokens": 9302211.0, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.04299815744161606, | |
| "eval_mean_token_accuracy": 0.9879275080319997, | |
| "eval_num_tokens": 9349372.0, | |
| "eval_runtime": 33.3905, | |
| "eval_samples_per_second": 11.051, | |
| "eval_steps_per_second": 5.541, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 4.036210018105009, | |
| "grad_norm": 0.06983848661184311, | |
| "learning_rate": 0.00012050063929608123, | |
| "loss": 0.0251, | |
| "mean_token_accuracy": 0.991089798126024, | |
| "num_tokens": 9446285.0, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 4.096560048280024, | |
| "grad_norm": 0.04937675595283508, | |
| "learning_rate": 0.0001164122074100633, | |
| "loss": 0.0204, | |
| "mean_token_accuracy": 0.9927009463310241, | |
| "num_tokens": 9582596.0, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.15691007845504, | |
| "grad_norm": 0.051658544689416885, | |
| "learning_rate": 0.00011234988267009415, | |
| "loss": 0.0238, | |
| "mean_token_accuracy": 0.9915762591362, | |
| "num_tokens": 9728310.0, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 4.217260108630055, | |
| "grad_norm": 0.09518906474113464, | |
| "learning_rate": 0.00010831682264316787, | |
| "loss": 0.0203, | |
| "mean_token_accuracy": 0.9925105625391006, | |
| "num_tokens": 9864477.0, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 4.27761013880507, | |
| "grad_norm": 0.09207163006067276, | |
| "learning_rate": 0.00010431616214937911, | |
| "loss": 0.0262, | |
| "mean_token_accuracy": 0.9906612086296082, | |
| "num_tokens": 10009271.0, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 4.337960168980085, | |
| "grad_norm": 0.037109535187482834, | |
| "learning_rate": 0.00010035101082528777, | |
| "loss": 0.0217, | |
| "mean_token_accuracy": 0.9920996063947678, | |
| "num_tokens": 10147200.0, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.3983101991551, | |
| "grad_norm": 0.06523178517818451, | |
| "learning_rate": 9.642445070685809e-05, | |
| "loss": 0.0243, | |
| "mean_token_accuracy": 0.9912169694900512, | |
| "num_tokens": 10292532.0, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 4.458660229330115, | |
| "grad_norm": 0.04405335709452629, | |
| "learning_rate": 9.253953383385157e-05, | |
| "loss": 0.0207, | |
| "mean_token_accuracy": 0.9923694771528244, | |
| "num_tokens": 10428880.0, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 4.51901025950513, | |
| "grad_norm": 0.05413525551557541, | |
| "learning_rate": 8.869927987753459e-05, | |
| "loss": 0.0238, | |
| "mean_token_accuracy": 0.9915979427099227, | |
| "num_tokens": 10574876.0, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 4.579360289680145, | |
| "grad_norm": 0.03870449215173721, | |
| "learning_rate": 8.490667379354661e-05, | |
| "loss": 0.0202, | |
| "mean_token_accuracy": 0.9924545711278916, | |
| "num_tokens": 10712888.0, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.63971031985516, | |
| "grad_norm": 0.07268328219652176, | |
| "learning_rate": 8.116466350175079e-05, | |
| "loss": 0.0257, | |
| "mean_token_accuracy": 0.9908116126060486, | |
| "num_tokens": 10859020.0, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 4.700060350030175, | |
| "grad_norm": 0.05169014260172844, | |
| "learning_rate": 7.747615759487304e-05, | |
| "loss": 0.0203, | |
| "mean_token_accuracy": 0.992445929646492, | |
| "num_tokens": 10996005.0, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 4.76041038020519, | |
| "grad_norm": 0.07525492459535599, | |
| "learning_rate": 7.38440230777085e-05, | |
| "loss": 0.0237, | |
| "mean_token_accuracy": 0.9914532697200775, | |
| "num_tokens": 11143151.0, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 4.820760410380205, | |
| "grad_norm": 0.09109634906053543, | |
| "learning_rate": 7.027108313865378e-05, | |
| "loss": 0.0204, | |
| "mean_token_accuracy": 0.9924772906303406, | |
| "num_tokens": 11279195.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.88111044055522, | |
| "grad_norm": 0.040585801005363464, | |
| "learning_rate": 6.676011495529687e-05, | |
| "loss": 0.0243, | |
| "mean_token_accuracy": 0.9914326167106629, | |
| "num_tokens": 11422435.0, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 4.941460470730235, | |
| "grad_norm": 0.04829643666744232, | |
| "learning_rate": 6.331384753577056e-05, | |
| "loss": 0.0199, | |
| "mean_token_accuracy": 0.992586697936058, | |
| "num_tokens": 11559063.0, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.08168815821409225, | |
| "learning_rate": 5.993495959754631e-05, | |
| "loss": 0.0216, | |
| "mean_token_accuracy": 0.9922702736461285, | |
| "num_tokens": 11686715.0, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.04248536005616188, | |
| "eval_mean_token_accuracy": 0.9885660554911639, | |
| "eval_num_tokens": 11686715.0, | |
| "eval_runtime": 33.36, | |
| "eval_samples_per_second": 11.061, | |
| "eval_steps_per_second": 5.546, | |
| "step": 2075 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 2905, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "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": 5.327805029835571e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |