| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 500, |
| "global_step": 250, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 2.0973353385925293, |
| "epoch": 0.004, |
| "grad_norm": 56.63088607788086, |
| "learning_rate": 0.0, |
| "loss": 8.81, |
| "mean_token_accuracy": 0.1589217633008957, |
| "num_tokens": 11020.0, |
| "step": 1 |
| }, |
| { |
| "entropy": 2.359926462173462, |
| "epoch": 0.008, |
| "grad_norm": 48.76058578491211, |
| "learning_rate": 2.5e-05, |
| "loss": 7.5185, |
| "mean_token_accuracy": 0.20284946262836456, |
| "num_tokens": 21059.0, |
| "step": 2 |
| }, |
| { |
| "entropy": 2.3848068714141846, |
| "epoch": 0.012, |
| "grad_norm": 51.160762786865234, |
| "learning_rate": 5e-05, |
| "loss": 8.4107, |
| "mean_token_accuracy": 0.1458083838224411, |
| "num_tokens": 31081.0, |
| "step": 3 |
| }, |
| { |
| "entropy": 2.5165743827819824, |
| "epoch": 0.016, |
| "grad_norm": 57.92226028442383, |
| "learning_rate": 7.500000000000001e-05, |
| "loss": 8.7345, |
| "mean_token_accuracy": 0.10784424096345901, |
| "num_tokens": 39994.0, |
| "step": 4 |
| }, |
| { |
| "entropy": 3.0336835384368896, |
| "epoch": 0.02, |
| "grad_norm": 42.479244232177734, |
| "learning_rate": 0.0001, |
| "loss": 6.6625, |
| "mean_token_accuracy": 0.15363158285617828, |
| "num_tokens": 49317.0, |
| "step": 5 |
| }, |
| { |
| "entropy": 3.2910828590393066, |
| "epoch": 0.024, |
| "grad_norm": 35.47465515136719, |
| "learning_rate": 0.000125, |
| "loss": 5.5297, |
| "mean_token_accuracy": 0.2194845825433731, |
| "num_tokens": 60805.0, |
| "step": 6 |
| }, |
| { |
| "entropy": 2.9953644275665283, |
| "epoch": 0.028, |
| "grad_norm": 24.23429298400879, |
| "learning_rate": 0.00015000000000000001, |
| "loss": 3.7, |
| "mean_token_accuracy": 0.40734514594078064, |
| "num_tokens": 72706.0, |
| "step": 7 |
| }, |
| { |
| "entropy": 2.88269305229187, |
| "epoch": 0.032, |
| "grad_norm": 25.993759155273438, |
| "learning_rate": 0.000175, |
| "loss": 2.7163, |
| "mean_token_accuracy": 0.5192126631736755, |
| "num_tokens": 85460.0, |
| "step": 8 |
| }, |
| { |
| "entropy": 2.2315902709960938, |
| "epoch": 0.036, |
| "grad_norm": 11.012210845947266, |
| "learning_rate": 0.0002, |
| "loss": 2.1933, |
| "mean_token_accuracy": 0.5909518003463745, |
| "num_tokens": 99299.0, |
| "step": 9 |
| }, |
| { |
| "entropy": 1.8418952226638794, |
| "epoch": 0.04, |
| "grad_norm": 5.978642463684082, |
| "learning_rate": 0.00019999241639982352, |
| "loss": 2.1315, |
| "mean_token_accuracy": 0.6005287170410156, |
| "num_tokens": 112919.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.6634621620178223, |
| "epoch": 0.044, |
| "grad_norm": 5.048736572265625, |
| "learning_rate": 0.00019996966687731613, |
| "loss": 2.092, |
| "mean_token_accuracy": 0.6170734763145447, |
| "num_tokens": 121074.0, |
| "step": 11 |
| }, |
| { |
| "entropy": 1.4140448570251465, |
| "epoch": 0.048, |
| "grad_norm": 3.720097303390503, |
| "learning_rate": 0.00019993175526632852, |
| "loss": 1.7816, |
| "mean_token_accuracy": 0.6572261452674866, |
| "num_tokens": 132057.0, |
| "step": 12 |
| }, |
| { |
| "entropy": 1.261852502822876, |
| "epoch": 0.052, |
| "grad_norm": 2.3393256664276123, |
| "learning_rate": 0.0001998786879558941, |
| "loss": 1.4815, |
| "mean_token_accuracy": 0.6984454989433289, |
| "num_tokens": 144153.0, |
| "step": 13 |
| }, |
| { |
| "entropy": 1.5389002561569214, |
| "epoch": 0.056, |
| "grad_norm": 1.971347451210022, |
| "learning_rate": 0.00019981047388915196, |
| "loss": 1.6526, |
| "mean_token_accuracy": 0.6537265777587891, |
| "num_tokens": 154768.0, |
| "step": 14 |
| }, |
| { |
| "entropy": 1.3573092222213745, |
| "epoch": 0.06, |
| "grad_norm": 1.3137863874435425, |
| "learning_rate": 0.00019972712456184022, |
| "loss": 1.3945, |
| "mean_token_accuracy": 0.6989641189575195, |
| "num_tokens": 165968.0, |
| "step": 15 |
| }, |
| { |
| "entropy": 1.477573037147522, |
| "epoch": 0.064, |
| "grad_norm": 1.0724811553955078, |
| "learning_rate": 0.00019962865402035814, |
| "loss": 1.5077, |
| "mean_token_accuracy": 0.6758458614349365, |
| "num_tokens": 176610.0, |
| "step": 16 |
| }, |
| { |
| "entropy": 1.062203288078308, |
| "epoch": 0.068, |
| "grad_norm": 0.7103755474090576, |
| "learning_rate": 0.00019951507885939952, |
| "loss": 1.1178, |
| "mean_token_accuracy": 0.7602011561393738, |
| "num_tokens": 188743.0, |
| "step": 17 |
| }, |
| { |
| "entropy": 1.2973206043243408, |
| "epoch": 0.072, |
| "grad_norm": 0.6334678530693054, |
| "learning_rate": 0.00019938641821915564, |
| "loss": 1.3933, |
| "mean_token_accuracy": 0.7131435871124268, |
| "num_tokens": 201253.0, |
| "step": 18 |
| }, |
| { |
| "entropy": 1.3070148229599, |
| "epoch": 0.076, |
| "grad_norm": 0.708212673664093, |
| "learning_rate": 0.00019924269378208997, |
| "loss": 1.3897, |
| "mean_token_accuracy": 0.6967664361000061, |
| "num_tokens": 211244.0, |
| "step": 19 |
| }, |
| { |
| "entropy": 1.4724856615066528, |
| "epoch": 0.08, |
| "grad_norm": 0.6635432839393616, |
| "learning_rate": 0.00019908392976928396, |
| "loss": 1.5634, |
| "mean_token_accuracy": 0.6638235449790955, |
| "num_tokens": 221446.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.0885224342346191, |
| "epoch": 0.084, |
| "grad_norm": 0.535370409488678, |
| "learning_rate": 0.0001989101529363553, |
| "loss": 1.1798, |
| "mean_token_accuracy": 0.7438462376594543, |
| "num_tokens": 232945.0, |
| "step": 21 |
| }, |
| { |
| "entropy": 1.2113245725631714, |
| "epoch": 0.088, |
| "grad_norm": 0.5345674753189087, |
| "learning_rate": 0.00019872139256894885, |
| "loss": 1.2599, |
| "mean_token_accuracy": 0.7245048880577087, |
| "num_tokens": 243753.0, |
| "step": 22 |
| }, |
| { |
| "entropy": 1.2911491394042969, |
| "epoch": 0.092, |
| "grad_norm": 0.5416544079780579, |
| "learning_rate": 0.00019851768047780144, |
| "loss": 1.3408, |
| "mean_token_accuracy": 0.7001619338989258, |
| "num_tokens": 256105.0, |
| "step": 23 |
| }, |
| { |
| "entropy": 1.128836750984192, |
| "epoch": 0.096, |
| "grad_norm": 0.48422297835350037, |
| "learning_rate": 0.00019829905099338086, |
| "loss": 1.1527, |
| "mean_token_accuracy": 0.748742938041687, |
| "num_tokens": 269432.0, |
| "step": 24 |
| }, |
| { |
| "entropy": 1.5081233978271484, |
| "epoch": 0.1, |
| "grad_norm": 0.5165230631828308, |
| "learning_rate": 0.00019806554096010035, |
| "loss": 1.4886, |
| "mean_token_accuracy": 0.680823802947998, |
| "num_tokens": 279582.0, |
| "step": 25 |
| }, |
| { |
| "entropy": 1.3273299932479858, |
| "epoch": 0.104, |
| "grad_norm": 0.534648597240448, |
| "learning_rate": 0.00019781718973010948, |
| "loss": 1.2789, |
| "mean_token_accuracy": 0.7176167964935303, |
| "num_tokens": 292021.0, |
| "step": 26 |
| }, |
| { |
| "entropy": 1.508842945098877, |
| "epoch": 0.108, |
| "grad_norm": 0.4762946665287018, |
| "learning_rate": 0.00019755403915666227, |
| "loss": 1.4154, |
| "mean_token_accuracy": 0.6926988959312439, |
| "num_tokens": 303035.0, |
| "step": 27 |
| }, |
| { |
| "entropy": 1.5553829669952393, |
| "epoch": 0.112, |
| "grad_norm": 0.5048977732658386, |
| "learning_rate": 0.00019727613358706396, |
| "loss": 1.4793, |
| "mean_token_accuracy": 0.683097779750824, |
| "num_tokens": 311456.0, |
| "step": 28 |
| }, |
| { |
| "entropy": 1.1076712608337402, |
| "epoch": 0.116, |
| "grad_norm": 0.40861934423446655, |
| "learning_rate": 0.00019698351985519735, |
| "loss": 1.0217, |
| "mean_token_accuracy": 0.7810451984405518, |
| "num_tokens": 323666.0, |
| "step": 29 |
| }, |
| { |
| "entropy": 1.2993760108947754, |
| "epoch": 0.12, |
| "grad_norm": 0.4528363049030304, |
| "learning_rate": 0.00019667624727363019, |
| "loss": 1.2028, |
| "mean_token_accuracy": 0.7430762052536011, |
| "num_tokens": 334753.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.3225066661834717, |
| "epoch": 0.124, |
| "grad_norm": 0.5367140173912048, |
| "learning_rate": 0.0001963543676253048, |
| "loss": 1.2189, |
| "mean_token_accuracy": 0.7315188050270081, |
| "num_tokens": 343683.0, |
| "step": 31 |
| }, |
| { |
| "entropy": 1.677242398262024, |
| "epoch": 0.128, |
| "grad_norm": 0.4724753201007843, |
| "learning_rate": 0.0001960179351548113, |
| "loss": 1.602, |
| "mean_token_accuracy": 0.6546955108642578, |
| "num_tokens": 357496.0, |
| "step": 32 |
| }, |
| { |
| "entropy": 0.9450194239616394, |
| "epoch": 0.132, |
| "grad_norm": 0.4643884003162384, |
| "learning_rate": 0.00019566700655924626, |
| "loss": 0.9468, |
| "mean_token_accuracy": 0.7905657291412354, |
| "num_tokens": 369306.0, |
| "step": 33 |
| }, |
| { |
| "entropy": 1.4320679903030396, |
| "epoch": 0.136, |
| "grad_norm": 0.4968578815460205, |
| "learning_rate": 0.00019530164097865768, |
| "loss": 1.4105, |
| "mean_token_accuracy": 0.6940935254096985, |
| "num_tokens": 381024.0, |
| "step": 34 |
| }, |
| { |
| "entropy": 1.4245866537094116, |
| "epoch": 0.14, |
| "grad_norm": 0.40614748001098633, |
| "learning_rate": 0.0001949218999860784, |
| "loss": 1.4487, |
| "mean_token_accuracy": 0.6868099570274353, |
| "num_tokens": 392891.0, |
| "step": 35 |
| }, |
| { |
| "entropy": 1.1022300720214844, |
| "epoch": 0.144, |
| "grad_norm": 0.3792162835597992, |
| "learning_rate": 0.00019452784757714974, |
| "loss": 1.1092, |
| "mean_token_accuracy": 0.761519730091095, |
| "num_tokens": 404395.0, |
| "step": 36 |
| }, |
| { |
| "entropy": 1.097092628479004, |
| "epoch": 0.148, |
| "grad_norm": 0.4267388880252838, |
| "learning_rate": 0.00019411955015933657, |
| "loss": 1.1114, |
| "mean_token_accuracy": 0.7575680613517761, |
| "num_tokens": 416190.0, |
| "step": 37 |
| }, |
| { |
| "entropy": 1.1763945817947388, |
| "epoch": 0.152, |
| "grad_norm": 0.49878817796707153, |
| "learning_rate": 0.00019369707654073601, |
| "loss": 1.2312, |
| "mean_token_accuracy": 0.7294224500656128, |
| "num_tokens": 428329.0, |
| "step": 38 |
| }, |
| { |
| "entropy": 1.103309154510498, |
| "epoch": 0.156, |
| "grad_norm": 0.5332479476928711, |
| "learning_rate": 0.00019326049791848154, |
| "loss": 1.0894, |
| "mean_token_accuracy": 0.7563456892967224, |
| "num_tokens": 438062.0, |
| "step": 39 |
| }, |
| { |
| "entropy": 1.3513190746307373, |
| "epoch": 0.16, |
| "grad_norm": 0.665856122970581, |
| "learning_rate": 0.00019280988786674467, |
| "loss": 1.4117, |
| "mean_token_accuracy": 0.6817158460617065, |
| "num_tokens": 446503.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.1249470710754395, |
| "epoch": 0.164, |
| "grad_norm": 0.44063600897789, |
| "learning_rate": 0.0001923453223243358, |
| "loss": 1.1345, |
| "mean_token_accuracy": 0.7451761960983276, |
| "num_tokens": 457544.0, |
| "step": 41 |
| }, |
| { |
| "entropy": 1.139631986618042, |
| "epoch": 0.168, |
| "grad_norm": 0.5003185272216797, |
| "learning_rate": 0.00019186687958190667, |
| "loss": 1.1346, |
| "mean_token_accuracy": 0.7483499050140381, |
| "num_tokens": 467091.0, |
| "step": 42 |
| }, |
| { |
| "entropy": 0.9406906962394714, |
| "epoch": 0.172, |
| "grad_norm": 0.6294145584106445, |
| "learning_rate": 0.00019137464026875654, |
| "loss": 0.9044, |
| "mean_token_accuracy": 0.7882066965103149, |
| "num_tokens": 475318.0, |
| "step": 43 |
| }, |
| { |
| "entropy": 1.039001703262329, |
| "epoch": 0.176, |
| "grad_norm": 0.4918760657310486, |
| "learning_rate": 0.0001908686873392441, |
| "loss": 1.0364, |
| "mean_token_accuracy": 0.7681604623794556, |
| "num_tokens": 485741.0, |
| "step": 44 |
| }, |
| { |
| "entropy": 1.1919207572937012, |
| "epoch": 0.18, |
| "grad_norm": 0.5181702971458435, |
| "learning_rate": 0.00019034910605880776, |
| "loss": 1.1167, |
| "mean_token_accuracy": 0.7450622916221619, |
| "num_tokens": 498907.0, |
| "step": 45 |
| }, |
| { |
| "entropy": 1.003733515739441, |
| "epoch": 0.184, |
| "grad_norm": 0.5143950581550598, |
| "learning_rate": 0.0001898159839895961, |
| "loss": 0.963, |
| "mean_token_accuracy": 0.7841158509254456, |
| "num_tokens": 510304.0, |
| "step": 46 |
| }, |
| { |
| "entropy": 1.3185124397277832, |
| "epoch": 0.188, |
| "grad_norm": 0.5639718174934387, |
| "learning_rate": 0.00018926941097571182, |
| "loss": 1.268, |
| "mean_token_accuracy": 0.7116304039955139, |
| "num_tokens": 521389.0, |
| "step": 47 |
| }, |
| { |
| "entropy": 1.1227295398712158, |
| "epoch": 0.192, |
| "grad_norm": 0.6089756488800049, |
| "learning_rate": 0.00018870947912807048, |
| "loss": 1.0814, |
| "mean_token_accuracy": 0.7578927874565125, |
| "num_tokens": 532097.0, |
| "step": 48 |
| }, |
| { |
| "entropy": 1.2459306716918945, |
| "epoch": 0.196, |
| "grad_norm": 0.7441679239273071, |
| "learning_rate": 0.00018813628280887795, |
| "loss": 1.2047, |
| "mean_token_accuracy": 0.7300441861152649, |
| "num_tokens": 541382.0, |
| "step": 49 |
| }, |
| { |
| "entropy": 1.0864311456680298, |
| "epoch": 0.2, |
| "grad_norm": 0.5735241174697876, |
| "learning_rate": 0.00018754991861572766, |
| "loss": 1.0827, |
| "mean_token_accuracy": 0.7548420429229736, |
| "num_tokens": 551039.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.9783708453178406, |
| "epoch": 0.204, |
| "grad_norm": 0.6659897565841675, |
| "learning_rate": 0.00018695048536532182, |
| "loss": 0.978, |
| "mean_token_accuracy": 0.7748624682426453, |
| "num_tokens": 562856.0, |
| "step": 51 |
| }, |
| { |
| "entropy": 1.0906152725219727, |
| "epoch": 0.208, |
| "grad_norm": 0.5842542052268982, |
| "learning_rate": 0.00018633808407681827, |
| "loss": 1.1066, |
| "mean_token_accuracy": 0.7530777454376221, |
| "num_tokens": 575692.0, |
| "step": 52 |
| }, |
| { |
| "entropy": 1.1353249549865723, |
| "epoch": 0.212, |
| "grad_norm": 1.0452736616134644, |
| "learning_rate": 0.0001857128179548063, |
| "loss": 1.2501, |
| "mean_token_accuracy": 0.7129384875297546, |
| "num_tokens": 587040.0, |
| "step": 53 |
| }, |
| { |
| "entropy": 1.1658884286880493, |
| "epoch": 0.216, |
| "grad_norm": 0.7569848299026489, |
| "learning_rate": 0.0001850747923719143, |
| "loss": 1.1789, |
| "mean_token_accuracy": 0.7339466214179993, |
| "num_tokens": 598130.0, |
| "step": 54 |
| }, |
| { |
| "entropy": 0.9339560866355896, |
| "epoch": 0.22, |
| "grad_norm": 0.6351683139801025, |
| "learning_rate": 0.0001844241148510516, |
| "loss": 0.9055, |
| "mean_token_accuracy": 0.7876133322715759, |
| "num_tokens": 611259.0, |
| "step": 55 |
| }, |
| { |
| "entropy": 1.1959530115127563, |
| "epoch": 0.224, |
| "grad_norm": 0.8705658912658691, |
| "learning_rate": 0.00018376089504728847, |
| "loss": 1.1617, |
| "mean_token_accuracy": 0.7372428774833679, |
| "num_tokens": 622686.0, |
| "step": 56 |
| }, |
| { |
| "entropy": 1.2461718320846558, |
| "epoch": 0.228, |
| "grad_norm": 0.8680962324142456, |
| "learning_rate": 0.0001830852447293765, |
| "loss": 1.1835, |
| "mean_token_accuracy": 0.7315577864646912, |
| "num_tokens": 632638.0, |
| "step": 57 |
| }, |
| { |
| "entropy": 1.1705292463302612, |
| "epoch": 0.232, |
| "grad_norm": 0.7268466353416443, |
| "learning_rate": 0.00018239727776091282, |
| "loss": 1.1152, |
| "mean_token_accuracy": 0.7414857745170593, |
| "num_tokens": 643974.0, |
| "step": 58 |
| }, |
| { |
| "entropy": 1.153640866279602, |
| "epoch": 0.236, |
| "grad_norm": 0.7480162978172302, |
| "learning_rate": 0.00018169711008115126, |
| "loss": 1.1153, |
| "mean_token_accuracy": 0.7440808415412903, |
| "num_tokens": 654366.0, |
| "step": 59 |
| }, |
| { |
| "entropy": 1.097459077835083, |
| "epoch": 0.24, |
| "grad_norm": 0.7212317585945129, |
| "learning_rate": 0.00018098485968546385, |
| "loss": 1.0895, |
| "mean_token_accuracy": 0.7482320666313171, |
| "num_tokens": 666105.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 1.05923593044281, |
| "epoch": 0.244, |
| "grad_norm": 0.8075670599937439, |
| "learning_rate": 0.00018026064660545552, |
| "loss": 1.067, |
| "mean_token_accuracy": 0.7528573870658875, |
| "num_tokens": 676781.0, |
| "step": 61 |
| }, |
| { |
| "entropy": 1.0399152040481567, |
| "epoch": 0.248, |
| "grad_norm": 0.9833256602287292, |
| "learning_rate": 0.00017952459288873623, |
| "loss": 1.0939, |
| "mean_token_accuracy": 0.7464929819107056, |
| "num_tokens": 686763.0, |
| "step": 62 |
| }, |
| { |
| "entropy": 1.0377105474472046, |
| "epoch": 0.252, |
| "grad_norm": 0.852209210395813, |
| "learning_rate": 0.00017877682257835242, |
| "loss": 1.0704, |
| "mean_token_accuracy": 0.7448949217796326, |
| "num_tokens": 700085.0, |
| "step": 63 |
| }, |
| { |
| "entropy": 0.8170187473297119, |
| "epoch": 0.256, |
| "grad_norm": 0.8792871832847595, |
| "learning_rate": 0.00017801746169188328, |
| "loss": 0.6947, |
| "mean_token_accuracy": 0.833350419998169, |
| "num_tokens": 709832.0, |
| "step": 64 |
| }, |
| { |
| "entropy": 1.0612165927886963, |
| "epoch": 0.26, |
| "grad_norm": 0.8374829888343811, |
| "learning_rate": 0.0001772466382002032, |
| "loss": 0.9753, |
| "mean_token_accuracy": 0.7734495401382446, |
| "num_tokens": 720847.0, |
| "step": 65 |
| }, |
| { |
| "entropy": 1.0511215925216675, |
| "epoch": 0.264, |
| "grad_norm": 0.8725335001945496, |
| "learning_rate": 0.00017646448200591596, |
| "loss": 1.0397, |
| "mean_token_accuracy": 0.7560197710990906, |
| "num_tokens": 732768.0, |
| "step": 66 |
| }, |
| { |
| "entropy": 1.16708242893219, |
| "epoch": 0.268, |
| "grad_norm": 0.9981763958930969, |
| "learning_rate": 0.0001756711249214627, |
| "loss": 1.2386, |
| "mean_token_accuracy": 0.7174420952796936, |
| "num_tokens": 743044.0, |
| "step": 67 |
| }, |
| { |
| "entropy": 0.9941532611846924, |
| "epoch": 0.272, |
| "grad_norm": 0.9159516096115112, |
| "learning_rate": 0.0001748667006469085, |
| "loss": 0.9769, |
| "mean_token_accuracy": 0.7736346125602722, |
| "num_tokens": 752915.0, |
| "step": 68 |
| }, |
| { |
| "entropy": 0.9163703322410583, |
| "epoch": 0.276, |
| "grad_norm": 0.7718108892440796, |
| "learning_rate": 0.0001740513447474104, |
| "loss": 0.8858, |
| "mean_token_accuracy": 0.79206782579422, |
| "num_tokens": 765070.0, |
| "step": 69 |
| }, |
| { |
| "entropy": 1.043192744255066, |
| "epoch": 0.28, |
| "grad_norm": 0.7897371649742126, |
| "learning_rate": 0.00017322519463037168, |
| "loss": 1.0291, |
| "mean_token_accuracy": 0.7658995985984802, |
| "num_tokens": 776110.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.8629508018493652, |
| "epoch": 0.284, |
| "grad_norm": 0.8085194230079651, |
| "learning_rate": 0.00017238838952228502, |
| "loss": 0.8871, |
| "mean_token_accuracy": 0.7935369610786438, |
| "num_tokens": 786850.0, |
| "step": 71 |
| }, |
| { |
| "entropy": 1.0944290161132812, |
| "epoch": 0.288, |
| "grad_norm": 0.9569735527038574, |
| "learning_rate": 0.00017154107044526964, |
| "loss": 1.0377, |
| "mean_token_accuracy": 0.7588126063346863, |
| "num_tokens": 796554.0, |
| "step": 72 |
| }, |
| { |
| "entropy": 0.7297062277793884, |
| "epoch": 0.292, |
| "grad_norm": 0.9553439021110535, |
| "learning_rate": 0.00017068338019330564, |
| "loss": 0.6367, |
| "mean_token_accuracy": 0.8526090979576111, |
| "num_tokens": 806387.0, |
| "step": 73 |
| }, |
| { |
| "entropy": 0.9341362714767456, |
| "epoch": 0.296, |
| "grad_norm": 0.9124618768692017, |
| "learning_rate": 0.00016981546330816953, |
| "loss": 0.8903, |
| "mean_token_accuracy": 0.7885122299194336, |
| "num_tokens": 816626.0, |
| "step": 74 |
| }, |
| { |
| "entropy": 0.5450064539909363, |
| "epoch": 0.3, |
| "grad_norm": 0.8968790769577026, |
| "learning_rate": 0.0001689374660550757, |
| "loss": 0.625, |
| "mean_token_accuracy": 0.8518457412719727, |
| "num_tokens": 828737.0, |
| "step": 75 |
| }, |
| { |
| "entropy": 0.7374886870384216, |
| "epoch": 0.304, |
| "grad_norm": 1.0030460357666016, |
| "learning_rate": 0.00016804953639802677, |
| "loss": 0.7191, |
| "mean_token_accuracy": 0.8276105523109436, |
| "num_tokens": 840068.0, |
| "step": 76 |
| }, |
| { |
| "entropy": 0.7412508130073547, |
| "epoch": 0.308, |
| "grad_norm": 1.2331219911575317, |
| "learning_rate": 0.00016715182397487844, |
| "loss": 0.8017, |
| "mean_token_accuracy": 0.8093528151512146, |
| "num_tokens": 851489.0, |
| "step": 77 |
| }, |
| { |
| "entropy": 1.0338571071624756, |
| "epoch": 0.312, |
| "grad_norm": 0.9198253154754639, |
| "learning_rate": 0.00016624448007212167, |
| "loss": 1.0164, |
| "mean_token_accuracy": 0.7691991329193115, |
| "num_tokens": 863640.0, |
| "step": 78 |
| }, |
| { |
| "entropy": 0.9462264776229858, |
| "epoch": 0.316, |
| "grad_norm": 0.7652691602706909, |
| "learning_rate": 0.0001653276575993871, |
| "loss": 0.9116, |
| "mean_token_accuracy": 0.7878068089485168, |
| "num_tokens": 878798.0, |
| "step": 79 |
| }, |
| { |
| "entropy": 1.003931999206543, |
| "epoch": 0.32, |
| "grad_norm": 0.9275044202804565, |
| "learning_rate": 0.0001644015110636762, |
| "loss": 0.9505, |
| "mean_token_accuracy": 0.774969220161438, |
| "num_tokens": 889363.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.708422839641571, |
| "epoch": 0.324, |
| "grad_norm": 0.9531024694442749, |
| "learning_rate": 0.00016346619654332298, |
| "loss": 0.6343, |
| "mean_token_accuracy": 0.8398601412773132, |
| "num_tokens": 899375.0, |
| "step": 81 |
| }, |
| { |
| "entropy": 1.028533935546875, |
| "epoch": 0.328, |
| "grad_norm": 0.9965426921844482, |
| "learning_rate": 0.00016252187166169107, |
| "loss": 0.9782, |
| "mean_token_accuracy": 0.7718014121055603, |
| "num_tokens": 908803.0, |
| "step": 82 |
| }, |
| { |
| "entropy": 0.7851366400718689, |
| "epoch": 0.332, |
| "grad_norm": 0.8076054453849792, |
| "learning_rate": 0.0001615686955606102, |
| "loss": 0.765, |
| "mean_token_accuracy": 0.8145228028297424, |
| "num_tokens": 920855.0, |
| "step": 83 |
| }, |
| { |
| "entropy": 0.9262005090713501, |
| "epoch": 0.336, |
| "grad_norm": 0.9758970737457275, |
| "learning_rate": 0.000160606828873557, |
| "loss": 0.9151, |
| "mean_token_accuracy": 0.7844375371932983, |
| "num_tokens": 931151.0, |
| "step": 84 |
| }, |
| { |
| "entropy": 0.6350160837173462, |
| "epoch": 0.34, |
| "grad_norm": 1.0767849683761597, |
| "learning_rate": 0.00015963643369858437, |
| "loss": 0.7041, |
| "mean_token_accuracy": 0.8317089080810547, |
| "num_tokens": 940387.0, |
| "step": 85 |
| }, |
| { |
| "entropy": 0.9434180855751038, |
| "epoch": 0.344, |
| "grad_norm": 1.190125823020935, |
| "learning_rate": 0.0001586576735710038, |
| "loss": 0.9578, |
| "mean_token_accuracy": 0.7776371240615845, |
| "num_tokens": 949869.0, |
| "step": 86 |
| }, |
| { |
| "entropy": 0.8093811273574829, |
| "epoch": 0.348, |
| "grad_norm": 0.9406972527503967, |
| "learning_rate": 0.00015767071343582588, |
| "loss": 0.7415, |
| "mean_token_accuracy": 0.8234021663665771, |
| "num_tokens": 959118.0, |
| "step": 87 |
| }, |
| { |
| "entropy": 0.7648224234580994, |
| "epoch": 0.352, |
| "grad_norm": 0.7738968729972839, |
| "learning_rate": 0.00015667571961996296, |
| "loss": 0.7306, |
| "mean_token_accuracy": 0.8252065181732178, |
| "num_tokens": 971466.0, |
| "step": 88 |
| }, |
| { |
| "entropy": 0.659711480140686, |
| "epoch": 0.356, |
| "grad_norm": 0.801194965839386, |
| "learning_rate": 0.000155672859804199, |
| "loss": 0.6591, |
| "mean_token_accuracy": 0.8405287265777588, |
| "num_tokens": 983119.0, |
| "step": 89 |
| }, |
| { |
| "entropy": 0.9075393080711365, |
| "epoch": 0.36, |
| "grad_norm": 1.1619821786880493, |
| "learning_rate": 0.00015466230299493125, |
| "loss": 0.8609, |
| "mean_token_accuracy": 0.7937808632850647, |
| "num_tokens": 992640.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.7225161790847778, |
| "epoch": 0.364, |
| "grad_norm": 0.9581741690635681, |
| "learning_rate": 0.00015364421949568853, |
| "loss": 0.688, |
| "mean_token_accuracy": 0.8366522192955017, |
| "num_tokens": 1003037.0, |
| "step": 91 |
| }, |
| { |
| "entropy": 0.7597986459732056, |
| "epoch": 0.368, |
| "grad_norm": 1.0615458488464355, |
| "learning_rate": 0.00015261878087843085, |
| "loss": 0.7226, |
| "mean_token_accuracy": 0.824986457824707, |
| "num_tokens": 1014101.0, |
| "step": 92 |
| }, |
| { |
| "entropy": 0.6454391479492188, |
| "epoch": 0.372, |
| "grad_norm": 0.9587013721466064, |
| "learning_rate": 0.00015158615995463566, |
| "loss": 0.6641, |
| "mean_token_accuracy": 0.8425232172012329, |
| "num_tokens": 1025311.0, |
| "step": 93 |
| }, |
| { |
| "entropy": 0.8424079418182373, |
| "epoch": 0.376, |
| "grad_norm": 1.2206848859786987, |
| "learning_rate": 0.00015054653074617443, |
| "loss": 0.9094, |
| "mean_token_accuracy": 0.7876355648040771, |
| "num_tokens": 1037024.0, |
| "step": 94 |
| }, |
| { |
| "entropy": 0.8714436292648315, |
| "epoch": 0.38, |
| "grad_norm": 1.2640056610107422, |
| "learning_rate": 0.00014950006845598603, |
| "loss": 0.8689, |
| "mean_token_accuracy": 0.7946674823760986, |
| "num_tokens": 1046815.0, |
| "step": 95 |
| }, |
| { |
| "entropy": 0.5807708501815796, |
| "epoch": 0.384, |
| "grad_norm": 0.8595563173294067, |
| "learning_rate": 0.00014844694943855067, |
| "loss": 0.5039, |
| "mean_token_accuracy": 0.8740711212158203, |
| "num_tokens": 1056775.0, |
| "step": 96 |
| }, |
| { |
| "entropy": 0.6972081661224365, |
| "epoch": 0.388, |
| "grad_norm": 0.8816290497779846, |
| "learning_rate": 0.00014738735117016978, |
| "loss": 0.6369, |
| "mean_token_accuracy": 0.8466954827308655, |
| "num_tokens": 1067429.0, |
| "step": 97 |
| }, |
| { |
| "entropy": 0.5604634881019592, |
| "epoch": 0.392, |
| "grad_norm": 0.9110663533210754, |
| "learning_rate": 0.0001463214522190571, |
| "loss": 0.501, |
| "mean_token_accuracy": 0.8726220726966858, |
| "num_tokens": 1078838.0, |
| "step": 98 |
| }, |
| { |
| "entropy": 0.8191442489624023, |
| "epoch": 0.396, |
| "grad_norm": 1.2301007509231567, |
| "learning_rate": 0.00014524943221524533, |
| "loss": 0.8723, |
| "mean_token_accuracy": 0.7879535555839539, |
| "num_tokens": 1089781.0, |
| "step": 99 |
| }, |
| { |
| "entropy": 0.9090349674224854, |
| "epoch": 0.4, |
| "grad_norm": 1.0221482515335083, |
| "learning_rate": 0.00014417147182031428, |
| "loss": 0.8904, |
| "mean_token_accuracy": 0.7897399067878723, |
| "num_tokens": 1102239.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.6675378084182739, |
| "epoch": 0.404, |
| "grad_norm": 1.0766475200653076, |
| "learning_rate": 0.0001430877526969448, |
| "loss": 0.5804, |
| "mean_token_accuracy": 0.8515596985816956, |
| "num_tokens": 1113397.0, |
| "step": 101 |
| }, |
| { |
| "entropy": 0.6440110802650452, |
| "epoch": 0.408, |
| "grad_norm": 0.9045482873916626, |
| "learning_rate": 0.00014199845747830446, |
| "loss": 0.6087, |
| "mean_token_accuracy": 0.8539875149726868, |
| "num_tokens": 1125261.0, |
| "step": 102 |
| }, |
| { |
| "entropy": 0.8418194651603699, |
| "epoch": 0.412, |
| "grad_norm": 1.4091824293136597, |
| "learning_rate": 0.00014090376973726896, |
| "loss": 0.9077, |
| "mean_token_accuracy": 0.7805910706520081, |
| "num_tokens": 1136598.0, |
| "step": 103 |
| }, |
| { |
| "entropy": 0.7668788433074951, |
| "epoch": 0.416, |
| "grad_norm": 1.2579201459884644, |
| "learning_rate": 0.00013980387395548608, |
| "loss": 0.8016, |
| "mean_token_accuracy": 0.8103581070899963, |
| "num_tokens": 1146988.0, |
| "step": 104 |
| }, |
| { |
| "entropy": 0.9310951232910156, |
| "epoch": 0.42, |
| "grad_norm": 1.0861616134643555, |
| "learning_rate": 0.0001386989554922857, |
| "loss": 0.9118, |
| "mean_token_accuracy": 0.7869442701339722, |
| "num_tokens": 1157269.0, |
| "step": 105 |
| }, |
| { |
| "entropy": 0.48249903321266174, |
| "epoch": 0.424, |
| "grad_norm": 1.08878755569458, |
| "learning_rate": 0.00013758920055344226, |
| "loss": 0.3668, |
| "mean_token_accuracy": 0.9031632542610168, |
| "num_tokens": 1167071.0, |
| "step": 106 |
| }, |
| { |
| "entropy": 0.689582884311676, |
| "epoch": 0.428, |
| "grad_norm": 1.249678611755371, |
| "learning_rate": 0.00013647479615979468, |
| "loss": 0.581, |
| "mean_token_accuracy": 0.8578310608863831, |
| "num_tokens": 1175584.0, |
| "step": 107 |
| }, |
| { |
| "entropy": 0.5986989736557007, |
| "epoch": 0.432, |
| "grad_norm": 0.9662414193153381, |
| "learning_rate": 0.0001353559301157287, |
| "loss": 0.5368, |
| "mean_token_accuracy": 0.8663188219070435, |
| "num_tokens": 1187278.0, |
| "step": 108 |
| }, |
| { |
| "entropy": 0.9165630340576172, |
| "epoch": 0.436, |
| "grad_norm": 1.785701036453247, |
| "learning_rate": 0.00013423279097752713, |
| "loss": 1.015, |
| "mean_token_accuracy": 0.7645655870437622, |
| "num_tokens": 1196085.0, |
| "step": 109 |
| }, |
| { |
| "entropy": 0.6729641556739807, |
| "epoch": 0.44, |
| "grad_norm": 1.1485984325408936, |
| "learning_rate": 0.0001331055680215937, |
| "loss": 0.6846, |
| "mean_token_accuracy": 0.8309648633003235, |
| "num_tokens": 1209457.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.5594127178192139, |
| "epoch": 0.444, |
| "grad_norm": 0.9643635153770447, |
| "learning_rate": 0.00013197445121255523, |
| "loss": 0.5453, |
| "mean_token_accuracy": 0.8717948794364929, |
| "num_tokens": 1219482.0, |
| "step": 111 |
| }, |
| { |
| "entropy": 0.8685882687568665, |
| "epoch": 0.448, |
| "grad_norm": 1.0649855136871338, |
| "learning_rate": 0.00013083963117124802, |
| "loss": 0.8078, |
| "mean_token_accuracy": 0.8030028939247131, |
| "num_tokens": 1232205.0, |
| "step": 112 |
| }, |
| { |
| "entropy": 0.6931924223899841, |
| "epoch": 0.452, |
| "grad_norm": 1.0093775987625122, |
| "learning_rate": 0.00012970129914259356, |
| "loss": 0.6155, |
| "mean_token_accuracy": 0.8536151647567749, |
| "num_tokens": 1242331.0, |
| "step": 113 |
| }, |
| { |
| "entropy": 0.940833568572998, |
| "epoch": 0.456, |
| "grad_norm": 1.1867045164108276, |
| "learning_rate": 0.00012855964696336896, |
| "loss": 0.9769, |
| "mean_token_accuracy": 0.7758588790893555, |
| "num_tokens": 1253219.0, |
| "step": 114 |
| }, |
| { |
| "entropy": 0.644160270690918, |
| "epoch": 0.46, |
| "grad_norm": 1.0405246019363403, |
| "learning_rate": 0.00012741486702987797, |
| "loss": 0.5985, |
| "mean_token_accuracy": 0.853334367275238, |
| "num_tokens": 1266012.0, |
| "step": 115 |
| }, |
| { |
| "entropy": 0.6467909812927246, |
| "epoch": 0.464, |
| "grad_norm": 1.0041617155075073, |
| "learning_rate": 0.00012626715226552745, |
| "loss": 0.589, |
| "mean_token_accuracy": 0.8532326817512512, |
| "num_tokens": 1278728.0, |
| "step": 116 |
| }, |
| { |
| "entropy": 0.6697924733161926, |
| "epoch": 0.468, |
| "grad_norm": 1.225751519203186, |
| "learning_rate": 0.00012511669608831485, |
| "loss": 0.6097, |
| "mean_token_accuracy": 0.8501642942428589, |
| "num_tokens": 1289989.0, |
| "step": 117 |
| }, |
| { |
| "entropy": 0.5410375595092773, |
| "epoch": 0.472, |
| "grad_norm": 1.1570745706558228, |
| "learning_rate": 0.00012396369237823286, |
| "loss": 0.4868, |
| "mean_token_accuracy": 0.8764288425445557, |
| "num_tokens": 1300139.0, |
| "step": 118 |
| }, |
| { |
| "entropy": 0.7634984254837036, |
| "epoch": 0.476, |
| "grad_norm": 1.26356041431427, |
| "learning_rate": 0.00012280833544459568, |
| "loss": 0.7205, |
| "mean_token_accuracy": 0.8254125118255615, |
| "num_tokens": 1310262.0, |
| "step": 119 |
| }, |
| { |
| "entropy": 0.7683166265487671, |
| "epoch": 0.48, |
| "grad_norm": 1.263581395149231, |
| "learning_rate": 0.0001216508199932932, |
| "loss": 0.8117, |
| "mean_token_accuracy": 0.8081841468811035, |
| "num_tokens": 1322385.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.7120152115821838, |
| "epoch": 0.484, |
| "grad_norm": 1.0431994199752808, |
| "learning_rate": 0.00012049134109397832, |
| "loss": 0.6529, |
| "mean_token_accuracy": 0.8445841073989868, |
| "num_tokens": 1337186.0, |
| "step": 121 |
| }, |
| { |
| "entropy": 0.7104185819625854, |
| "epoch": 0.488, |
| "grad_norm": 1.0483670234680176, |
| "learning_rate": 0.00011933009414719301, |
| "loss": 0.7264, |
| "mean_token_accuracy": 0.8267722725868225, |
| "num_tokens": 1349686.0, |
| "step": 122 |
| }, |
| { |
| "entropy": 0.9004740118980408, |
| "epoch": 0.492, |
| "grad_norm": 1.2728326320648193, |
| "learning_rate": 0.0001181672748514383, |
| "loss": 0.8288, |
| "mean_token_accuracy": 0.7991729378700256, |
| "num_tokens": 1361295.0, |
| "step": 123 |
| }, |
| { |
| "entropy": 0.7324482798576355, |
| "epoch": 0.496, |
| "grad_norm": 1.113187313079834, |
| "learning_rate": 0.00011700307917019457, |
| "loss": 0.6708, |
| "mean_token_accuracy": 0.8380022048950195, |
| "num_tokens": 1371328.0, |
| "step": 124 |
| }, |
| { |
| "entropy": 0.4651513993740082, |
| "epoch": 0.5, |
| "grad_norm": 0.9615445733070374, |
| "learning_rate": 0.00011583770329889687, |
| "loss": 0.3927, |
| "mean_token_accuracy": 0.8962194919586182, |
| "num_tokens": 1383577.0, |
| "step": 125 |
| }, |
| { |
| "entropy": 0.7203623056411743, |
| "epoch": 0.504, |
| "grad_norm": 1.9821685552597046, |
| "learning_rate": 0.00011467134363187099, |
| "loss": 0.7477, |
| "mean_token_accuracy": 0.8203157782554626, |
| "num_tokens": 1393079.0, |
| "step": 126 |
| }, |
| { |
| "entropy": 0.6460074186325073, |
| "epoch": 0.508, |
| "grad_norm": 2.170309543609619, |
| "learning_rate": 0.0001135041967292364, |
| "loss": 0.7293, |
| "mean_token_accuracy": 0.8246410489082336, |
| "num_tokens": 1402553.0, |
| "step": 127 |
| }, |
| { |
| "entropy": 0.5284869074821472, |
| "epoch": 0.512, |
| "grad_norm": 1.1976059675216675, |
| "learning_rate": 0.00011233645928378105, |
| "loss": 0.4515, |
| "mean_token_accuracy": 0.8835397362709045, |
| "num_tokens": 1412567.0, |
| "step": 128 |
| }, |
| { |
| "entropy": 0.4933440089225769, |
| "epoch": 0.516, |
| "grad_norm": 1.259605050086975, |
| "learning_rate": 0.00011116832808781378, |
| "loss": 0.4273, |
| "mean_token_accuracy": 0.8937007784843445, |
| "num_tokens": 1424507.0, |
| "step": 129 |
| }, |
| { |
| "entropy": 0.7313682436943054, |
| "epoch": 0.52, |
| "grad_norm": 1.3257278203964233, |
| "learning_rate": 0.00011000000000000002, |
| "loss": 0.6628, |
| "mean_token_accuracy": 0.8386083245277405, |
| "num_tokens": 1433419.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.4719448387622833, |
| "epoch": 0.524, |
| "grad_norm": 1.0190857648849487, |
| "learning_rate": 0.00010883167191218624, |
| "loss": 0.394, |
| "mean_token_accuracy": 0.9015878438949585, |
| "num_tokens": 1444568.0, |
| "step": 131 |
| }, |
| { |
| "entropy": 0.671485185623169, |
| "epoch": 0.528, |
| "grad_norm": 1.0616258382797241, |
| "learning_rate": 0.00010766354071621898, |
| "loss": 0.6002, |
| "mean_token_accuracy": 0.8531094193458557, |
| "num_tokens": 1458029.0, |
| "step": 132 |
| }, |
| { |
| "entropy": 0.37650296092033386, |
| "epoch": 0.532, |
| "grad_norm": 1.3387216329574585, |
| "learning_rate": 0.00010649580327076363, |
| "loss": 0.3286, |
| "mean_token_accuracy": 0.9205968379974365, |
| "num_tokens": 1469290.0, |
| "step": 133 |
| }, |
| { |
| "entropy": 0.5341144800186157, |
| "epoch": 0.536, |
| "grad_norm": 1.3382490873336792, |
| "learning_rate": 0.00010532865636812906, |
| "loss": 0.5154, |
| "mean_token_accuracy": 0.8731343150138855, |
| "num_tokens": 1482290.0, |
| "step": 134 |
| }, |
| { |
| "entropy": 0.5627509951591492, |
| "epoch": 0.54, |
| "grad_norm": 1.9341037273406982, |
| "learning_rate": 0.00010416229670110314, |
| "loss": 0.6165, |
| "mean_token_accuracy": 0.8522540330886841, |
| "num_tokens": 1491409.0, |
| "step": 135 |
| }, |
| { |
| "entropy": 0.40396589040756226, |
| "epoch": 0.544, |
| "grad_norm": 1.1782172918319702, |
| "learning_rate": 0.0001029969208298054, |
| "loss": 0.3003, |
| "mean_token_accuracy": 0.9240846037864685, |
| "num_tokens": 1501106.0, |
| "step": 136 |
| }, |
| { |
| "entropy": 0.7882451415061951, |
| "epoch": 0.548, |
| "grad_norm": 1.2071242332458496, |
| "learning_rate": 0.00010183272514856172, |
| "loss": 0.7402, |
| "mean_token_accuracy": 0.8205325603485107, |
| "num_tokens": 1511511.0, |
| "step": 137 |
| }, |
| { |
| "entropy": 0.645850419998169, |
| "epoch": 0.552, |
| "grad_norm": 1.2258726358413696, |
| "learning_rate": 0.00010066990585280704, |
| "loss": 0.6628, |
| "mean_token_accuracy": 0.8416417241096497, |
| "num_tokens": 1523208.0, |
| "step": 138 |
| }, |
| { |
| "entropy": 0.42336830496788025, |
| "epoch": 0.556, |
| "grad_norm": 1.5242387056350708, |
| "learning_rate": 9.95086589060217e-05, |
| "loss": 0.2851, |
| "mean_token_accuracy": 0.9278303980827332, |
| "num_tokens": 1532078.0, |
| "step": 139 |
| }, |
| { |
| "entropy": 0.7408692240715027, |
| "epoch": 0.56, |
| "grad_norm": 1.1541547775268555, |
| "learning_rate": 9.834918000670681e-05, |
| "loss": 0.6626, |
| "mean_token_accuracy": 0.8373293280601501, |
| "num_tokens": 1544092.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.5522950887680054, |
| "epoch": 0.564, |
| "grad_norm": 1.2949033975601196, |
| "learning_rate": 9.719166455540436e-05, |
| "loss": 0.4288, |
| "mean_token_accuracy": 0.8930450081825256, |
| "num_tokens": 1554360.0, |
| "step": 141 |
| }, |
| { |
| "entropy": 0.6674742698669434, |
| "epoch": 0.568, |
| "grad_norm": 1.4206823110580444, |
| "learning_rate": 9.603630762176718e-05, |
| "loss": 0.6521, |
| "mean_token_accuracy": 0.8428549766540527, |
| "num_tokens": 1567420.0, |
| "step": 142 |
| }, |
| { |
| "entropy": 0.48524826765060425, |
| "epoch": 0.572, |
| "grad_norm": 2.079292058944702, |
| "learning_rate": 9.488330391168516e-05, |
| "loss": 0.5301, |
| "mean_token_accuracy": 0.8732667565345764, |
| "num_tokens": 1577735.0, |
| "step": 143 |
| }, |
| { |
| "entropy": 0.6761812567710876, |
| "epoch": 0.576, |
| "grad_norm": 2.401937246322632, |
| "learning_rate": 9.373284773447259e-05, |
| "loss": 0.7367, |
| "mean_token_accuracy": 0.8284353613853455, |
| "num_tokens": 1587605.0, |
| "step": 144 |
| }, |
| { |
| "entropy": 0.6450383067131042, |
| "epoch": 0.58, |
| "grad_norm": 1.4243282079696655, |
| "learning_rate": 9.258513297012204e-05, |
| "loss": 0.5896, |
| "mean_token_accuracy": 0.8545143008232117, |
| "num_tokens": 1598694.0, |
| "step": 145 |
| }, |
| { |
| "entropy": 0.6930188536643982, |
| "epoch": 0.584, |
| "grad_norm": 1.256988763809204, |
| "learning_rate": 9.14403530366311e-05, |
| "loss": 0.6308, |
| "mean_token_accuracy": 0.8471713662147522, |
| "num_tokens": 1610716.0, |
| "step": 146 |
| }, |
| { |
| "entropy": 0.5218924283981323, |
| "epoch": 0.588, |
| "grad_norm": 1.4413528442382812, |
| "learning_rate": 9.029870085740649e-05, |
| "loss": 0.3976, |
| "mean_token_accuracy": 0.8993197083473206, |
| "num_tokens": 1621743.0, |
| "step": 147 |
| }, |
| { |
| "entropy": 0.6881991028785706, |
| "epoch": 0.592, |
| "grad_norm": 1.6851704120635986, |
| "learning_rate": 8.916036882875198e-05, |
| "loss": 0.5696, |
| "mean_token_accuracy": 0.8667510151863098, |
| "num_tokens": 1635186.0, |
| "step": 148 |
| }, |
| { |
| "entropy": 0.5879225134849548, |
| "epoch": 0.596, |
| "grad_norm": 1.0631791353225708, |
| "learning_rate": 8.802554878744481e-05, |
| "loss": 0.5184, |
| "mean_token_accuracy": 0.8734765648841858, |
| "num_tokens": 1647660.0, |
| "step": 149 |
| }, |
| { |
| "entropy": 0.2675434947013855, |
| "epoch": 0.6, |
| "grad_norm": 0.8148598670959473, |
| "learning_rate": 8.689443197840636e-05, |
| "loss": 0.1821, |
| "mean_token_accuracy": 0.9514302015304565, |
| "num_tokens": 1658080.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.7003239989280701, |
| "epoch": 0.604, |
| "grad_norm": 2.1313421726226807, |
| "learning_rate": 8.576720902247291e-05, |
| "loss": 0.7331, |
| "mean_token_accuracy": 0.8224857449531555, |
| "num_tokens": 1668622.0, |
| "step": 151 |
| }, |
| { |
| "entropy": 0.3838817775249481, |
| "epoch": 0.608, |
| "grad_norm": 1.2568340301513672, |
| "learning_rate": 8.464406988427134e-05, |
| "loss": 0.3143, |
| "mean_token_accuracy": 0.9187853336334229, |
| "num_tokens": 1678536.0, |
| "step": 152 |
| }, |
| { |
| "entropy": 0.5554172992706299, |
| "epoch": 0.612, |
| "grad_norm": 2.047208786010742, |
| "learning_rate": 8.352520384020535e-05, |
| "loss": 0.5846, |
| "mean_token_accuracy": 0.8602444529533386, |
| "num_tokens": 1690974.0, |
| "step": 153 |
| }, |
| { |
| "entropy": 0.45566004514694214, |
| "epoch": 0.616, |
| "grad_norm": 1.319417953491211, |
| "learning_rate": 8.241079944655776e-05, |
| "loss": 0.429, |
| "mean_token_accuracy": 0.8949615955352783, |
| "num_tokens": 1702686.0, |
| "step": 154 |
| }, |
| { |
| "entropy": 0.592678964138031, |
| "epoch": 0.62, |
| "grad_norm": 1.4309483766555786, |
| "learning_rate": 8.130104450771434e-05, |
| "loss": 0.5834, |
| "mean_token_accuracy": 0.8569459319114685, |
| "num_tokens": 1712845.0, |
| "step": 155 |
| }, |
| { |
| "entropy": 0.6900731921195984, |
| "epoch": 0.624, |
| "grad_norm": 1.4166367053985596, |
| "learning_rate": 8.019612604451394e-05, |
| "loss": 0.5856, |
| "mean_token_accuracy": 0.8577383160591125, |
| "num_tokens": 1723883.0, |
| "step": 156 |
| }, |
| { |
| "entropy": 0.6701188087463379, |
| "epoch": 0.628, |
| "grad_norm": 1.5286020040512085, |
| "learning_rate": 7.909623026273107e-05, |
| "loss": 0.5537, |
| "mean_token_accuracy": 0.8609554767608643, |
| "num_tokens": 1733702.0, |
| "step": 157 |
| }, |
| { |
| "entropy": 0.6067744493484497, |
| "epoch": 0.632, |
| "grad_norm": 1.2828212976455688, |
| "learning_rate": 7.80015425216956e-05, |
| "loss": 0.5559, |
| "mean_token_accuracy": 0.8636751770973206, |
| "num_tokens": 1744245.0, |
| "step": 158 |
| }, |
| { |
| "entropy": 0.510319173336029, |
| "epoch": 0.636, |
| "grad_norm": 1.2530028820037842, |
| "learning_rate": 7.69122473030552e-05, |
| "loss": 0.4339, |
| "mean_token_accuracy": 0.8923482894897461, |
| "num_tokens": 1755617.0, |
| "step": 159 |
| }, |
| { |
| "entropy": 0.5396801829338074, |
| "epoch": 0.64, |
| "grad_norm": 1.4599785804748535, |
| "learning_rate": 7.582852817968576e-05, |
| "loss": 0.4212, |
| "mean_token_accuracy": 0.8964771628379822, |
| "num_tokens": 1766718.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.5178655385971069, |
| "epoch": 0.644, |
| "grad_norm": 1.1682523488998413, |
| "learning_rate": 7.475056778475469e-05, |
| "loss": 0.4751, |
| "mean_token_accuracy": 0.8846276998519897, |
| "num_tokens": 1776107.0, |
| "step": 161 |
| }, |
| { |
| "entropy": 0.48712846636772156, |
| "epoch": 0.648, |
| "grad_norm": 1.3003511428833008, |
| "learning_rate": 7.367854778094291e-05, |
| "loss": 0.4654, |
| "mean_token_accuracy": 0.8929698467254639, |
| "num_tokens": 1786564.0, |
| "step": 162 |
| }, |
| { |
| "entropy": 0.3912014961242676, |
| "epoch": 0.652, |
| "grad_norm": 1.110151767730713, |
| "learning_rate": 7.261264882983024e-05, |
| "loss": 0.3429, |
| "mean_token_accuracy": 0.915362536907196, |
| "num_tokens": 1798771.0, |
| "step": 163 |
| }, |
| { |
| "entropy": 0.55037522315979, |
| "epoch": 0.656, |
| "grad_norm": 1.7164385318756104, |
| "learning_rate": 7.155305056144937e-05, |
| "loss": 0.5739, |
| "mean_token_accuracy": 0.8635889291763306, |
| "num_tokens": 1811228.0, |
| "step": 164 |
| }, |
| { |
| "entropy": 0.47252118587493896, |
| "epoch": 0.66, |
| "grad_norm": 1.010840654373169, |
| "learning_rate": 7.0499931544014e-05, |
| "loss": 0.4453, |
| "mean_token_accuracy": 0.890230119228363, |
| "num_tokens": 1822745.0, |
| "step": 165 |
| }, |
| { |
| "entropy": 0.4593927264213562, |
| "epoch": 0.664, |
| "grad_norm": 1.1940021514892578, |
| "learning_rate": 6.94534692538256e-05, |
| "loss": 0.3447, |
| "mean_token_accuracy": 0.9129361510276794, |
| "num_tokens": 1834497.0, |
| "step": 166 |
| }, |
| { |
| "entropy": 0.5545728802680969, |
| "epoch": 0.668, |
| "grad_norm": 1.5106139183044434, |
| "learning_rate": 6.841384004536434e-05, |
| "loss": 0.5047, |
| "mean_token_accuracy": 0.8748116493225098, |
| "num_tokens": 1841800.0, |
| "step": 167 |
| }, |
| { |
| "entropy": 0.4316740036010742, |
| "epoch": 0.672, |
| "grad_norm": 1.1023913621902466, |
| "learning_rate": 6.738121912156914e-05, |
| "loss": 0.3654, |
| "mean_token_accuracy": 0.9105084538459778, |
| "num_tokens": 1855077.0, |
| "step": 168 |
| }, |
| { |
| "entropy": 0.5366147756576538, |
| "epoch": 0.676, |
| "grad_norm": 1.2672346830368042, |
| "learning_rate": 6.63557805043115e-05, |
| "loss": 0.4416, |
| "mean_token_accuracy": 0.8936067819595337, |
| "num_tokens": 1865465.0, |
| "step": 169 |
| }, |
| { |
| "entropy": 0.5674976110458374, |
| "epoch": 0.68, |
| "grad_norm": 1.43003249168396, |
| "learning_rate": 6.533769700506875e-05, |
| "loss": 0.4565, |
| "mean_token_accuracy": 0.8876559138298035, |
| "num_tokens": 1875650.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.49446940422058105, |
| "epoch": 0.684, |
| "grad_norm": 1.2005430459976196, |
| "learning_rate": 6.4327140195801e-05, |
| "loss": 0.417, |
| "mean_token_accuracy": 0.9007508158683777, |
| "num_tokens": 1885375.0, |
| "step": 171 |
| }, |
| { |
| "entropy": 0.6743717193603516, |
| "epoch": 0.688, |
| "grad_norm": 1.4010748863220215, |
| "learning_rate": 6.332428038003706e-05, |
| "loss": 0.637, |
| "mean_token_accuracy": 0.8471227884292603, |
| "num_tokens": 1896412.0, |
| "step": 172 |
| }, |
| { |
| "entropy": 0.49895045161247253, |
| "epoch": 0.692, |
| "grad_norm": 1.7875854969024658, |
| "learning_rate": 6.232928656417416e-05, |
| "loss": 0.449, |
| "mean_token_accuracy": 0.8918893933296204, |
| "num_tokens": 1907116.0, |
| "step": 173 |
| }, |
| { |
| "entropy": 0.3326154351234436, |
| "epoch": 0.696, |
| "grad_norm": 0.9512577652931213, |
| "learning_rate": 6.134232642899626e-05, |
| "loss": 0.2697, |
| "mean_token_accuracy": 0.9238321781158447, |
| "num_tokens": 1917201.0, |
| "step": 174 |
| }, |
| { |
| "entropy": 0.4969325065612793, |
| "epoch": 0.7, |
| "grad_norm": 1.3267816305160522, |
| "learning_rate": 6.036356630141565e-05, |
| "loss": 0.4614, |
| "mean_token_accuracy": 0.8914347887039185, |
| "num_tokens": 1929205.0, |
| "step": 175 |
| }, |
| { |
| "entropy": 0.5694164037704468, |
| "epoch": 0.704, |
| "grad_norm": 1.334541916847229, |
| "learning_rate": 5.9393171126443005e-05, |
| "loss": 0.5125, |
| "mean_token_accuracy": 0.8812792301177979, |
| "num_tokens": 1942246.0, |
| "step": 176 |
| }, |
| { |
| "entropy": 0.658866286277771, |
| "epoch": 0.708, |
| "grad_norm": 1.2868497371673584, |
| "learning_rate": 5.843130443938983e-05, |
| "loss": 0.6569, |
| "mean_token_accuracy": 0.8416849374771118, |
| "num_tokens": 1956397.0, |
| "step": 177 |
| }, |
| { |
| "entropy": 0.7168298959732056, |
| "epoch": 0.712, |
| "grad_norm": 1.1754109859466553, |
| "learning_rate": 5.747812833830895e-05, |
| "loss": 0.7006, |
| "mean_token_accuracy": 0.8340783715248108, |
| "num_tokens": 1967808.0, |
| "step": 178 |
| }, |
| { |
| "entropy": 0.7076151967048645, |
| "epoch": 0.716, |
| "grad_norm": 1.5503113269805908, |
| "learning_rate": 5.653380345667705e-05, |
| "loss": 0.616, |
| "mean_token_accuracy": 0.8446282148361206, |
| "num_tokens": 1980869.0, |
| "step": 179 |
| }, |
| { |
| "entropy": 0.7145153880119324, |
| "epoch": 0.72, |
| "grad_norm": 1.6007428169250488, |
| "learning_rate": 5.559848893632385e-05, |
| "loss": 0.6182, |
| "mean_token_accuracy": 0.8411088585853577, |
| "num_tokens": 1992703.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.698581337928772, |
| "epoch": 0.724, |
| "grad_norm": 1.5277796983718872, |
| "learning_rate": 5.467234240061293e-05, |
| "loss": 0.6196, |
| "mean_token_accuracy": 0.8507580757141113, |
| "num_tokens": 2001543.0, |
| "step": 181 |
| }, |
| { |
| "entropy": 0.5171747803688049, |
| "epoch": 0.728, |
| "grad_norm": 1.39127516746521, |
| "learning_rate": 5.375551992787835e-05, |
| "loss": 0.3984, |
| "mean_token_accuracy": 0.9049353003501892, |
| "num_tokens": 2011980.0, |
| "step": 182 |
| }, |
| { |
| "entropy": 0.509817898273468, |
| "epoch": 0.732, |
| "grad_norm": 1.0949788093566895, |
| "learning_rate": 5.2848176025121535e-05, |
| "loss": 0.4335, |
| "mean_token_accuracy": 0.8921334147453308, |
| "num_tokens": 2024516.0, |
| "step": 183 |
| }, |
| { |
| "entropy": 0.4842955470085144, |
| "epoch": 0.736, |
| "grad_norm": 1.3567887544631958, |
| "learning_rate": 5.195046360197325e-05, |
| "loss": 0.4199, |
| "mean_token_accuracy": 0.8933181762695312, |
| "num_tokens": 2035054.0, |
| "step": 184 |
| }, |
| { |
| "entropy": 0.48027482628822327, |
| "epoch": 0.74, |
| "grad_norm": 1.9636229276657104, |
| "learning_rate": 5.106253394492435e-05, |
| "loss": 0.4542, |
| "mean_token_accuracy": 0.8890615105628967, |
| "num_tokens": 2044710.0, |
| "step": 185 |
| }, |
| { |
| "entropy": 0.47678428888320923, |
| "epoch": 0.744, |
| "grad_norm": 1.8577853441238403, |
| "learning_rate": 5.0184536691830476e-05, |
| "loss": 0.4531, |
| "mean_token_accuracy": 0.8908271789550781, |
| "num_tokens": 2056922.0, |
| "step": 186 |
| }, |
| { |
| "entropy": 0.4041728079319, |
| "epoch": 0.748, |
| "grad_norm": 1.4495080709457397, |
| "learning_rate": 4.931661980669439e-05, |
| "loss": 0.3819, |
| "mean_token_accuracy": 0.9036104083061218, |
| "num_tokens": 2068834.0, |
| "step": 187 |
| }, |
| { |
| "entropy": 0.3090624511241913, |
| "epoch": 0.752, |
| "grad_norm": 1.0853344202041626, |
| "learning_rate": 4.845892955473038e-05, |
| "loss": 0.291, |
| "mean_token_accuracy": 0.9313488006591797, |
| "num_tokens": 2080795.0, |
| "step": 188 |
| }, |
| { |
| "entropy": 0.5363933444023132, |
| "epoch": 0.756, |
| "grad_norm": 1.2786738872528076, |
| "learning_rate": 4.7611610477715026e-05, |
| "loss": 0.5352, |
| "mean_token_accuracy": 0.867972731590271, |
| "num_tokens": 2093711.0, |
| "step": 189 |
| }, |
| { |
| "entropy": 0.48970481753349304, |
| "epoch": 0.76, |
| "grad_norm": 1.483815312385559, |
| "learning_rate": 4.6774805369628324e-05, |
| "loss": 0.3646, |
| "mean_token_accuracy": 0.9098596572875977, |
| "num_tokens": 2104474.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.5898410081863403, |
| "epoch": 0.764, |
| "grad_norm": 1.2790271043777466, |
| "learning_rate": 4.59486552525896e-05, |
| "loss": 0.5362, |
| "mean_token_accuracy": 0.8709191083908081, |
| "num_tokens": 2114431.0, |
| "step": 191 |
| }, |
| { |
| "entropy": 0.7715857028961182, |
| "epoch": 0.768, |
| "grad_norm": 1.4260997772216797, |
| "learning_rate": 4.51332993530915e-05, |
| "loss": 0.688, |
| "mean_token_accuracy": 0.8307946920394897, |
| "num_tokens": 2125922.0, |
| "step": 192 |
| }, |
| { |
| "entropy": 0.5289206504821777, |
| "epoch": 0.772, |
| "grad_norm": 1.214869499206543, |
| "learning_rate": 4.43288750785373e-05, |
| "loss": 0.4668, |
| "mean_token_accuracy": 0.8906829953193665, |
| "num_tokens": 2136700.0, |
| "step": 193 |
| }, |
| { |
| "entropy": 0.4115394949913025, |
| "epoch": 0.776, |
| "grad_norm": 1.061026930809021, |
| "learning_rate": 4.3535517994084064e-05, |
| "loss": 0.333, |
| "mean_token_accuracy": 0.9176801443099976, |
| "num_tokens": 2148789.0, |
| "step": 194 |
| }, |
| { |
| "entropy": 0.4777131974697113, |
| "epoch": 0.78, |
| "grad_norm": 1.1228642463684082, |
| "learning_rate": 4.275336179979683e-05, |
| "loss": 0.4044, |
| "mean_token_accuracy": 0.9039993286132812, |
| "num_tokens": 2161093.0, |
| "step": 195 |
| }, |
| { |
| "entropy": 0.42037472128868103, |
| "epoch": 0.784, |
| "grad_norm": 1.2192764282226562, |
| "learning_rate": 4.198253830811677e-05, |
| "loss": 0.3361, |
| "mean_token_accuracy": 0.9074494242668152, |
| "num_tokens": 2172116.0, |
| "step": 196 |
| }, |
| { |
| "entropy": 0.549899160861969, |
| "epoch": 0.788, |
| "grad_norm": 1.4125242233276367, |
| "learning_rate": 4.12231774216476e-05, |
| "loss": 0.4924, |
| "mean_token_accuracy": 0.879365086555481, |
| "num_tokens": 2181883.0, |
| "step": 197 |
| }, |
| { |
| "entropy": 0.42284148931503296, |
| "epoch": 0.792, |
| "grad_norm": 1.1463818550109863, |
| "learning_rate": 4.0475407111263817e-05, |
| "loss": 0.3531, |
| "mean_token_accuracy": 0.9119821190834045, |
| "num_tokens": 2194871.0, |
| "step": 198 |
| }, |
| { |
| "entropy": 0.5124052166938782, |
| "epoch": 0.796, |
| "grad_norm": 1.7839992046356201, |
| "learning_rate": 3.9739353394544477e-05, |
| "loss": 0.4233, |
| "mean_token_accuracy": 0.898973286151886, |
| "num_tokens": 2204613.0, |
| "step": 199 |
| }, |
| { |
| "entropy": 0.46030405163764954, |
| "epoch": 0.8, |
| "grad_norm": 1.3598767518997192, |
| "learning_rate": 3.901514031453619e-05, |
| "loss": 0.4143, |
| "mean_token_accuracy": 0.9042191505432129, |
| "num_tokens": 2215494.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.5470673441886902, |
| "epoch": 0.804, |
| "grad_norm": 1.4430607557296753, |
| "learning_rate": 3.830288991884876e-05, |
| "loss": 0.527, |
| "mean_token_accuracy": 0.8680225610733032, |
| "num_tokens": 2224543.0, |
| "step": 201 |
| }, |
| { |
| "entropy": 0.43841931223869324, |
| "epoch": 0.808, |
| "grad_norm": 1.203965187072754, |
| "learning_rate": 3.7602722239087204e-05, |
| "loss": 0.4137, |
| "mean_token_accuracy": 0.894135594367981, |
| "num_tokens": 2236362.0, |
| "step": 202 |
| }, |
| { |
| "entropy": 0.4509918987751007, |
| "epoch": 0.812, |
| "grad_norm": 1.2725450992584229, |
| "learning_rate": 3.691475527062352e-05, |
| "loss": 0.4209, |
| "mean_token_accuracy": 0.8999525308609009, |
| "num_tokens": 2246899.0, |
| "step": 203 |
| }, |
| { |
| "entropy": 0.5433053970336914, |
| "epoch": 0.816, |
| "grad_norm": 1.3696599006652832, |
| "learning_rate": 3.6239104952711543e-05, |
| "loss": 0.463, |
| "mean_token_accuracy": 0.8843159079551697, |
| "num_tokens": 2254992.0, |
| "step": 204 |
| }, |
| { |
| "entropy": 0.5255707502365112, |
| "epoch": 0.82, |
| "grad_norm": 1.202747106552124, |
| "learning_rate": 3.557588514894843e-05, |
| "loss": 0.453, |
| "mean_token_accuracy": 0.8915954232215881, |
| "num_tokens": 2268213.0, |
| "step": 205 |
| }, |
| { |
| "entropy": 0.5911931395530701, |
| "epoch": 0.824, |
| "grad_norm": 1.4482736587524414, |
| "learning_rate": 3.492520762808572e-05, |
| "loss": 0.4987, |
| "mean_token_accuracy": 0.8807631134986877, |
| "num_tokens": 2279537.0, |
| "step": 206 |
| }, |
| { |
| "entropy": 0.4150068759918213, |
| "epoch": 0.828, |
| "grad_norm": 1.2278997898101807, |
| "learning_rate": 3.428718204519369e-05, |
| "loss": 0.2919, |
| "mean_token_accuracy": 0.9287025928497314, |
| "num_tokens": 2292611.0, |
| "step": 207 |
| }, |
| { |
| "entropy": 0.3332892954349518, |
| "epoch": 0.832, |
| "grad_norm": 1.3437390327453613, |
| "learning_rate": 3.366191592318176e-05, |
| "loss": 0.2095, |
| "mean_token_accuracy": 0.9491567015647888, |
| "num_tokens": 2304886.0, |
| "step": 208 |
| }, |
| { |
| "entropy": 0.6497487425804138, |
| "epoch": 0.836, |
| "grad_norm": 1.673365831375122, |
| "learning_rate": 3.3049514634678204e-05, |
| "loss": 0.6501, |
| "mean_token_accuracy": 0.8420344591140747, |
| "num_tokens": 2313067.0, |
| "step": 209 |
| }, |
| { |
| "entropy": 0.6357022523880005, |
| "epoch": 0.84, |
| "grad_norm": 1.2789956331253052, |
| "learning_rate": 3.245008138427235e-05, |
| "loss": 0.5881, |
| "mean_token_accuracy": 0.8542210459709167, |
| "num_tokens": 2327248.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.5384138226509094, |
| "epoch": 0.844, |
| "grad_norm": 1.6131337881088257, |
| "learning_rate": 3.186371719112206e-05, |
| "loss": 0.5152, |
| "mean_token_accuracy": 0.8732239007949829, |
| "num_tokens": 2339918.0, |
| "step": 211 |
| }, |
| { |
| "entropy": 0.5150956511497498, |
| "epoch": 0.848, |
| "grad_norm": 1.2851109504699707, |
| "learning_rate": 3.1290520871929525e-05, |
| "loss": 0.4407, |
| "mean_token_accuracy": 0.8878676295280457, |
| "num_tokens": 2350800.0, |
| "step": 212 |
| }, |
| { |
| "entropy": 0.5567672252655029, |
| "epoch": 0.852, |
| "grad_norm": 1.4837950468063354, |
| "learning_rate": 3.0730589024288226e-05, |
| "loss": 0.4587, |
| "mean_token_accuracy": 0.8859366774559021, |
| "num_tokens": 2364163.0, |
| "step": 213 |
| }, |
| { |
| "entropy": 0.5855469107627869, |
| "epoch": 0.856, |
| "grad_norm": 1.4510327577590942, |
| "learning_rate": 3.0184016010403938e-05, |
| "loss": 0.5231, |
| "mean_token_accuracy": 0.8731569647789001, |
| "num_tokens": 2373389.0, |
| "step": 214 |
| }, |
| { |
| "entropy": 0.42299529910087585, |
| "epoch": 0.86, |
| "grad_norm": 1.2330726385116577, |
| "learning_rate": 2.965089394119227e-05, |
| "loss": 0.332, |
| "mean_token_accuracy": 0.9186156392097473, |
| "num_tokens": 2387116.0, |
| "step": 215 |
| }, |
| { |
| "entropy": 0.43762826919555664, |
| "epoch": 0.864, |
| "grad_norm": 1.1993961334228516, |
| "learning_rate": 2.9131312660755904e-05, |
| "loss": 0.3706, |
| "mean_token_accuracy": 0.9101641178131104, |
| "num_tokens": 2399062.0, |
| "step": 216 |
| }, |
| { |
| "entropy": 0.5866155624389648, |
| "epoch": 0.868, |
| "grad_norm": 1.4377245903015137, |
| "learning_rate": 2.8625359731243486e-05, |
| "loss": 0.4993, |
| "mean_token_accuracy": 0.8804880380630493, |
| "num_tokens": 2410703.0, |
| "step": 217 |
| }, |
| { |
| "entropy": 0.5209600329399109, |
| "epoch": 0.872, |
| "grad_norm": 1.2498382329940796, |
| "learning_rate": 2.8133120418093347e-05, |
| "loss": 0.4693, |
| "mean_token_accuracy": 0.8875941634178162, |
| "num_tokens": 2420660.0, |
| "step": 218 |
| }, |
| { |
| "entropy": 0.4361816346645355, |
| "epoch": 0.876, |
| "grad_norm": 1.188899040222168, |
| "learning_rate": 2.765467767566422e-05, |
| "loss": 0.3418, |
| "mean_token_accuracy": 0.9192692637443542, |
| "num_tokens": 2430460.0, |
| "step": 219 |
| }, |
| { |
| "entropy": 0.5995680093765259, |
| "epoch": 0.88, |
| "grad_norm": 1.4681471586227417, |
| "learning_rate": 2.7190112133255337e-05, |
| "loss": 0.5184, |
| "mean_token_accuracy": 0.8720136284828186, |
| "num_tokens": 2439838.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.5519295930862427, |
| "epoch": 0.884, |
| "grad_norm": 1.2246594429016113, |
| "learning_rate": 2.673950208151847e-05, |
| "loss": 0.477, |
| "mean_token_accuracy": 0.8802193403244019, |
| "num_tokens": 2450234.0, |
| "step": 221 |
| }, |
| { |
| "entropy": 0.6201585531234741, |
| "epoch": 0.888, |
| "grad_norm": 1.4758093357086182, |
| "learning_rate": 2.6302923459264006e-05, |
| "loss": 0.5529, |
| "mean_token_accuracy": 0.8612440228462219, |
| "num_tokens": 2460686.0, |
| "step": 222 |
| }, |
| { |
| "entropy": 0.5428203344345093, |
| "epoch": 0.892, |
| "grad_norm": 1.0732289552688599, |
| "learning_rate": 2.5880449840663413e-05, |
| "loss": 0.4952, |
| "mean_token_accuracy": 0.8789449334144592, |
| "num_tokens": 2473806.0, |
| "step": 223 |
| }, |
| { |
| "entropy": 0.5736177563667297, |
| "epoch": 0.896, |
| "grad_norm": 1.1641875505447388, |
| "learning_rate": 2.547215242285026e-05, |
| "loss": 0.5083, |
| "mean_token_accuracy": 0.8738900423049927, |
| "num_tokens": 2486646.0, |
| "step": 224 |
| }, |
| { |
| "entropy": 0.3208327293395996, |
| "epoch": 0.9, |
| "grad_norm": 1.3467357158660889, |
| "learning_rate": 2.507810001392163e-05, |
| "loss": 0.2289, |
| "mean_token_accuracy": 0.9426351189613342, |
| "num_tokens": 2495399.0, |
| "step": 225 |
| }, |
| { |
| "entropy": 0.43385663628578186, |
| "epoch": 0.904, |
| "grad_norm": 1.3956007957458496, |
| "learning_rate": 2.469835902134236e-05, |
| "loss": 0.3405, |
| "mean_token_accuracy": 0.9122757315635681, |
| "num_tokens": 2505934.0, |
| "step": 226 |
| }, |
| { |
| "entropy": 0.399286150932312, |
| "epoch": 0.908, |
| "grad_norm": 1.0955601930618286, |
| "learning_rate": 2.4332993440753746e-05, |
| "loss": 0.322, |
| "mean_token_accuracy": 0.9209378957748413, |
| "num_tokens": 2518091.0, |
| "step": 227 |
| }, |
| { |
| "entropy": 0.5215837955474854, |
| "epoch": 0.912, |
| "grad_norm": 1.1155186891555786, |
| "learning_rate": 2.398206484518871e-05, |
| "loss": 0.4673, |
| "mean_token_accuracy": 0.8811452984809875, |
| "num_tokens": 2532657.0, |
| "step": 228 |
| }, |
| { |
| "entropy": 0.5278834104537964, |
| "epoch": 0.916, |
| "grad_norm": 1.7340788841247559, |
| "learning_rate": 2.3645632374695246e-05, |
| "loss": 0.4788, |
| "mean_token_accuracy": 0.8789379596710205, |
| "num_tokens": 2542150.0, |
| "step": 229 |
| }, |
| { |
| "entropy": 0.47815102338790894, |
| "epoch": 0.92, |
| "grad_norm": 1.61054265499115, |
| "learning_rate": 2.3323752726369835e-05, |
| "loss": 0.442, |
| "mean_token_accuracy": 0.8940832614898682, |
| "num_tokens": 2554473.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.4165140390396118, |
| "epoch": 0.924, |
| "grad_norm": 1.14963698387146, |
| "learning_rate": 2.3016480144802675e-05, |
| "loss": 0.328, |
| "mean_token_accuracy": 0.9183060526847839, |
| "num_tokens": 2564794.0, |
| "step": 231 |
| }, |
| { |
| "entropy": 0.38186386227607727, |
| "epoch": 0.928, |
| "grad_norm": 1.187225580215454, |
| "learning_rate": 2.272386641293606e-05, |
| "loss": 0.3076, |
| "mean_token_accuracy": 0.9227842688560486, |
| "num_tokens": 2574522.0, |
| "step": 232 |
| }, |
| { |
| "entropy": 0.46965232491493225, |
| "epoch": 0.932, |
| "grad_norm": 1.2771713733673096, |
| "learning_rate": 2.2445960843337744e-05, |
| "loss": 0.3789, |
| "mean_token_accuracy": 0.9045142531394958, |
| "num_tokens": 2587793.0, |
| "step": 233 |
| }, |
| { |
| "entropy": 0.39393627643585205, |
| "epoch": 0.936, |
| "grad_norm": 1.0977147817611694, |
| "learning_rate": 2.218281026989053e-05, |
| "loss": 0.2874, |
| "mean_token_accuracy": 0.933529257774353, |
| "num_tokens": 2597664.0, |
| "step": 234 |
| }, |
| { |
| "entropy": 0.4673425555229187, |
| "epoch": 0.94, |
| "grad_norm": 1.0229376554489136, |
| "learning_rate": 2.1934459039899662e-05, |
| "loss": 0.3796, |
| "mean_token_accuracy": 0.9101356267929077, |
| "num_tokens": 2611676.0, |
| "step": 235 |
| }, |
| { |
| "entropy": 0.3621935546398163, |
| "epoch": 0.944, |
| "grad_norm": 1.1729012727737427, |
| "learning_rate": 2.1700949006619168e-05, |
| "loss": 0.2125, |
| "mean_token_accuracy": 0.9498928785324097, |
| "num_tokens": 2622415.0, |
| "step": 236 |
| }, |
| { |
| "entropy": 0.47119855880737305, |
| "epoch": 0.948, |
| "grad_norm": 1.2110202312469482, |
| "learning_rate": 2.1482319522198588e-05, |
| "loss": 0.3581, |
| "mean_token_accuracy": 0.9117594957351685, |
| "num_tokens": 2633693.0, |
| "step": 237 |
| }, |
| { |
| "entropy": 0.6427318453788757, |
| "epoch": 0.952, |
| "grad_norm": 1.4238137006759644, |
| "learning_rate": 2.127860743105119e-05, |
| "loss": 0.6113, |
| "mean_token_accuracy": 0.8502632975578308, |
| "num_tokens": 2646798.0, |
| "step": 238 |
| }, |
| { |
| "entropy": 0.6078838109970093, |
| "epoch": 0.956, |
| "grad_norm": 1.5493922233581543, |
| "learning_rate": 2.108984706364474e-05, |
| "loss": 0.5309, |
| "mean_token_accuracy": 0.874631941318512, |
| "num_tokens": 2657329.0, |
| "step": 239 |
| }, |
| { |
| "entropy": 0.4323933720588684, |
| "epoch": 0.96, |
| "grad_norm": 1.1048017740249634, |
| "learning_rate": 2.0916070230716063e-05, |
| "loss": 0.3455, |
| "mean_token_accuracy": 0.9157264232635498, |
| "num_tokens": 2668853.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.6867014169692993, |
| "epoch": 0.964, |
| "grad_norm": 1.5176868438720703, |
| "learning_rate": 2.0757306217910032e-05, |
| "loss": 0.7058, |
| "mean_token_accuracy": 0.8293052911758423, |
| "num_tokens": 2680068.0, |
| "step": 241 |
| }, |
| { |
| "entropy": 0.4175410568714142, |
| "epoch": 0.968, |
| "grad_norm": 1.427336573600769, |
| "learning_rate": 2.0613581780844353e-05, |
| "loss": 0.3294, |
| "mean_token_accuracy": 0.9166304469108582, |
| "num_tokens": 2689270.0, |
| "step": 242 |
| }, |
| { |
| "entropy": 0.30933189392089844, |
| "epoch": 0.972, |
| "grad_norm": 1.0740463733673096, |
| "learning_rate": 2.048492114060048e-05, |
| "loss": 0.2382, |
| "mean_token_accuracy": 0.9348580837249756, |
| "num_tokens": 2701077.0, |
| "step": 243 |
| }, |
| { |
| "entropy": 0.4613304138183594, |
| "epoch": 0.976, |
| "grad_norm": 1.318914532661438, |
| "learning_rate": 2.037134597964185e-05, |
| "loss": 0.4104, |
| "mean_token_accuracy": 0.8936565518379211, |
| "num_tokens": 2711846.0, |
| "step": 244 |
| }, |
| { |
| "entropy": 0.4268723130226135, |
| "epoch": 0.98, |
| "grad_norm": 1.142288327217102, |
| "learning_rate": 2.0272875438159798e-05, |
| "loss": 0.3691, |
| "mean_token_accuracy": 0.9119612574577332, |
| "num_tokens": 2725467.0, |
| "step": 245 |
| }, |
| { |
| "entropy": 0.5718749165534973, |
| "epoch": 0.984, |
| "grad_norm": 1.352400541305542, |
| "learning_rate": 2.018952611084803e-05, |
| "loss": 0.4803, |
| "mean_token_accuracy": 0.8806823492050171, |
| "num_tokens": 2735962.0, |
| "step": 246 |
| }, |
| { |
| "entropy": 0.4914339780807495, |
| "epoch": 0.988, |
| "grad_norm": 1.3557024002075195, |
| "learning_rate": 2.012131204410593e-05, |
| "loss": 0.4089, |
| "mean_token_accuracy": 0.8950228095054626, |
| "num_tokens": 2745166.0, |
| "step": 247 |
| }, |
| { |
| "entropy": 0.3986694812774658, |
| "epoch": 0.992, |
| "grad_norm": 1.19772469997406, |
| "learning_rate": 2.0068244733671478e-05, |
| "loss": 0.283, |
| "mean_token_accuracy": 0.9307823777198792, |
| "num_tokens": 2757477.0, |
| "step": 248 |
| }, |
| { |
| "entropy": 0.49750185012817383, |
| "epoch": 0.996, |
| "grad_norm": 1.2758872509002686, |
| "learning_rate": 2.0030333122683887e-05, |
| "loss": 0.4291, |
| "mean_token_accuracy": 0.8971313834190369, |
| "num_tokens": 2767414.0, |
| "step": 249 |
| }, |
| { |
| "entropy": 0.41299375891685486, |
| "epoch": 1.0, |
| "grad_norm": 1.3803913593292236, |
| "learning_rate": 2.0007583600176482e-05, |
| "loss": 0.3336, |
| "mean_token_accuracy": 0.9214777946472168, |
| "num_tokens": 2777133.0, |
| "step": 250 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 250, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "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": 3.78349455918081e+17, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
