| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.49930871025083945, | |
| "eval_steps": 16, | |
| "global_step": 79, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.006320363420896702, | |
| "grad_norm": 0.48046875, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6497, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.006320363420896702, | |
| "eval_loss": 0.5998793244361877, | |
| "eval_runtime": 56.2331, | |
| "eval_samples_per_second": 18.957, | |
| "eval_steps_per_second": 18.957, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.012640726841793404, | |
| "grad_norm": 0.470703125, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6349, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.018961090262690106, | |
| "grad_norm": 0.46875, | |
| "learning_rate": 6e-05, | |
| "loss": 0.5832, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.025281453683586808, | |
| "grad_norm": 0.47265625, | |
| "learning_rate": 8e-05, | |
| "loss": 0.557, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.03160181710448351, | |
| "grad_norm": 0.35546875, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4966, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03792218052538021, | |
| "grad_norm": 0.31640625, | |
| "learning_rate": 0.00012, | |
| "loss": 0.3771, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.04424254394627691, | |
| "grad_norm": 0.298828125, | |
| "learning_rate": 0.00014, | |
| "loss": 0.318, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.050562907367173615, | |
| "grad_norm": 0.36328125, | |
| "learning_rate": 0.00016, | |
| "loss": 0.296, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.05688327078807032, | |
| "grad_norm": 0.3046875, | |
| "learning_rate": 0.00018, | |
| "loss": 0.2682, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.06320363420896702, | |
| "grad_norm": 0.29296875, | |
| "learning_rate": 0.0002, | |
| "loss": 0.3039, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06952399762986372, | |
| "grad_norm": 0.1845703125, | |
| "learning_rate": 0.00019999770790755575, | |
| "loss": 0.3096, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.07584436105076042, | |
| "grad_norm": 0.16796875, | |
| "learning_rate": 0.00019999083173529673, | |
| "loss": 0.2575, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08216472447165712, | |
| "grad_norm": 0.1689453125, | |
| "learning_rate": 0.00019997937179843937, | |
| "loss": 0.289, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.08848508789255383, | |
| "grad_norm": 0.1572265625, | |
| "learning_rate": 0.0001999633286223284, | |
| "loss": 0.2878, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.09480545131345053, | |
| "grad_norm": 0.12353515625, | |
| "learning_rate": 0.00019994270294241266, | |
| "loss": 0.2274, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.10112581473434723, | |
| "grad_norm": 0.11572265625, | |
| "learning_rate": 0.00019991749570421146, | |
| "loss": 0.2252, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.10112581473434723, | |
| "eval_loss": 0.25023654103279114, | |
| "eval_runtime": 56.1603, | |
| "eval_samples_per_second": 18.981, | |
| "eval_steps_per_second": 18.981, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.10744617815524393, | |
| "grad_norm": 0.1298828125, | |
| "learning_rate": 0.0001998877080632712, | |
| "loss": 0.2512, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.11376654157614063, | |
| "grad_norm": 0.193359375, | |
| "learning_rate": 0.00019985334138511237, | |
| "loss": 0.2659, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.12008690499703734, | |
| "grad_norm": 0.142578125, | |
| "learning_rate": 0.00019981439724516716, | |
| "loss": 0.2415, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.12640726841793404, | |
| "grad_norm": 0.126953125, | |
| "learning_rate": 0.0001997708774287068, | |
| "loss": 0.2661, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13272763183883074, | |
| "grad_norm": 0.1142578125, | |
| "learning_rate": 0.00019972278393076023, | |
| "loss": 0.2046, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.13904799525972744, | |
| "grad_norm": 0.11865234375, | |
| "learning_rate": 0.0001996701189560223, | |
| "loss": 0.2087, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.14536835868062414, | |
| "grad_norm": 0.1435546875, | |
| "learning_rate": 0.00019961288491875278, | |
| "loss": 0.2246, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.15168872210152085, | |
| "grad_norm": 0.119140625, | |
| "learning_rate": 0.00019955108444266585, | |
| "loss": 0.1831, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.15800908552241755, | |
| "grad_norm": 0.1376953125, | |
| "learning_rate": 0.00019948472036080949, | |
| "loss": 0.2398, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.16432944894331425, | |
| "grad_norm": 0.1201171875, | |
| "learning_rate": 0.00019941379571543596, | |
| "loss": 0.189, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.17064981236421095, | |
| "grad_norm": 0.12890625, | |
| "learning_rate": 0.00019933831375786216, | |
| "loss": 0.2156, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.17697017578510765, | |
| "grad_norm": 0.1259765625, | |
| "learning_rate": 0.00019925827794832056, | |
| "loss": 0.2012, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.18329053920600435, | |
| "grad_norm": 0.11572265625, | |
| "learning_rate": 0.00019917369195580063, | |
| "loss": 0.1602, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.18961090262690106, | |
| "grad_norm": 0.12890625, | |
| "learning_rate": 0.00019908455965788067, | |
| "loss": 0.1976, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19593126604779776, | |
| "grad_norm": 0.11279296875, | |
| "learning_rate": 0.00019899088514055004, | |
| "loss": 0.1874, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.20225162946869446, | |
| "grad_norm": 0.1357421875, | |
| "learning_rate": 0.00019889267269802176, | |
| "loss": 0.2024, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.20225162946869446, | |
| "eval_loss": 0.2020280808210373, | |
| "eval_runtime": 56.2737, | |
| "eval_samples_per_second": 18.943, | |
| "eval_steps_per_second": 18.943, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.20857199288959116, | |
| "grad_norm": 0.1123046875, | |
| "learning_rate": 0.00019878992683253582, | |
| "loss": 0.1819, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.21489235631048786, | |
| "grad_norm": 0.1123046875, | |
| "learning_rate": 0.00019868265225415265, | |
| "loss": 0.1794, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.22121271973138457, | |
| "grad_norm": 0.12060546875, | |
| "learning_rate": 0.00019857085388053723, | |
| "loss": 0.1943, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.22753308315228127, | |
| "grad_norm": 0.134765625, | |
| "learning_rate": 0.00019845453683673368, | |
| "loss": 0.2265, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.23385344657317797, | |
| "grad_norm": 0.115234375, | |
| "learning_rate": 0.00019833370645493047, | |
| "loss": 0.181, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.24017380999407467, | |
| "grad_norm": 0.12353515625, | |
| "learning_rate": 0.0001982083682742156, | |
| "loss": 0.1946, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.24649417341497137, | |
| "grad_norm": 0.107421875, | |
| "learning_rate": 0.00019807852804032305, | |
| "loss": 0.1418, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.2528145368358681, | |
| "grad_norm": 0.11572265625, | |
| "learning_rate": 0.00019794419170536916, | |
| "loss": 0.1651, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2591349002567648, | |
| "grad_norm": 0.1337890625, | |
| "learning_rate": 0.00019780536542758, | |
| "loss": 0.1821, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.2654552636776615, | |
| "grad_norm": 0.12353515625, | |
| "learning_rate": 0.00019766205557100868, | |
| "loss": 0.1913, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.2717756270985582, | |
| "grad_norm": 0.12451171875, | |
| "learning_rate": 0.00019751426870524407, | |
| "loss": 0.194, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.2780959905194549, | |
| "grad_norm": 0.1162109375, | |
| "learning_rate": 0.00019736201160510931, | |
| "loss": 0.1832, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.2844163539403516, | |
| "grad_norm": 0.12353515625, | |
| "learning_rate": 0.0001972052912503514, | |
| "loss": 0.19, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2907367173612483, | |
| "grad_norm": 0.11474609375, | |
| "learning_rate": 0.00019704411482532116, | |
| "loss": 0.1875, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.297057080782145, | |
| "grad_norm": 0.1123046875, | |
| "learning_rate": 0.00019687848971864389, | |
| "loss": 0.184, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.3033774442030417, | |
| "grad_norm": 0.107421875, | |
| "learning_rate": 0.0001967084235228807, | |
| "loss": 0.1581, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3033774442030417, | |
| "eval_loss": 0.18037649989128113, | |
| "eval_runtime": 56.5509, | |
| "eval_samples_per_second": 18.85, | |
| "eval_steps_per_second": 18.85, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3096978076239384, | |
| "grad_norm": 0.10302734375, | |
| "learning_rate": 0.00019653392403418043, | |
| "loss": 0.1766, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.3160181710448351, | |
| "grad_norm": 0.11669921875, | |
| "learning_rate": 0.0001963549992519223, | |
| "loss": 0.1888, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3223385344657318, | |
| "grad_norm": 0.10009765625, | |
| "learning_rate": 0.00019617165737834916, | |
| "loss": 0.139, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.3286588978866285, | |
| "grad_norm": 0.12451171875, | |
| "learning_rate": 0.0001959839068181914, | |
| "loss": 0.1845, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.3349792613075252, | |
| "grad_norm": 0.12060546875, | |
| "learning_rate": 0.00019579175617828187, | |
| "loss": 0.2043, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.3412996247284219, | |
| "grad_norm": 0.1142578125, | |
| "learning_rate": 0.00019559521426716118, | |
| "loss": 0.1678, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.3476199881493186, | |
| "grad_norm": 0.11669921875, | |
| "learning_rate": 0.0001953942900946739, | |
| "loss": 0.1671, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.3539403515702153, | |
| "grad_norm": 0.115234375, | |
| "learning_rate": 0.00019518899287155556, | |
| "loss": 0.1724, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.360260714991112, | |
| "grad_norm": 0.126953125, | |
| "learning_rate": 0.0001949793320090105, | |
| "loss": 0.177, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.3665810784120087, | |
| "grad_norm": 0.1259765625, | |
| "learning_rate": 0.00019476531711828027, | |
| "loss": 0.1644, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.3729014418329054, | |
| "grad_norm": 0.1142578125, | |
| "learning_rate": 0.0001945469580102031, | |
| "loss": 0.1564, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.3792218052538021, | |
| "grad_norm": 0.10302734375, | |
| "learning_rate": 0.0001943242646947643, | |
| "loss": 0.1353, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3855421686746988, | |
| "grad_norm": 0.10791015625, | |
| "learning_rate": 0.00019409724738063714, | |
| "loss": 0.1622, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.3918625320955955, | |
| "grad_norm": 0.10791015625, | |
| "learning_rate": 0.00019386591647471506, | |
| "loss": 0.1564, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.3981828955164922, | |
| "grad_norm": 0.11376953125, | |
| "learning_rate": 0.00019363028258163447, | |
| "loss": 0.176, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.4045032589373889, | |
| "grad_norm": 0.1162109375, | |
| "learning_rate": 0.00019339035650328869, | |
| "loss": 0.1912, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.4045032589373889, | |
| "eval_loss": 0.1681559830904007, | |
| "eval_runtime": 56.4912, | |
| "eval_samples_per_second": 18.87, | |
| "eval_steps_per_second": 18.87, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.4108236223582856, | |
| "grad_norm": 0.11572265625, | |
| "learning_rate": 0.0001931461492383327, | |
| "loss": 0.1959, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.4171439857791823, | |
| "grad_norm": 0.0966796875, | |
| "learning_rate": 0.00019289767198167916, | |
| "loss": 0.1379, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.423464349200079, | |
| "grad_norm": 0.1142578125, | |
| "learning_rate": 0.00019264493612398481, | |
| "loss": 0.1669, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.42978471262097573, | |
| "grad_norm": 0.09619140625, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.1279, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.43610507604187243, | |
| "grad_norm": 0.10986328125, | |
| "learning_rate": 0.0001921267351436808, | |
| "loss": 0.1535, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.44242543946276913, | |
| "grad_norm": 0.11962890625, | |
| "learning_rate": 0.0001918612937763622, | |
| "loss": 0.1697, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.44874580288366583, | |
| "grad_norm": 0.107421875, | |
| "learning_rate": 0.00019159164131749587, | |
| "loss": 0.166, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.45506616630456254, | |
| "grad_norm": 0.10986328125, | |
| "learning_rate": 0.00019131779012844912, | |
| "loss": 0.1508, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.46138652972545924, | |
| "grad_norm": 0.1201171875, | |
| "learning_rate": 0.00019103975276306678, | |
| "loss": 0.1617, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.46770689314635594, | |
| "grad_norm": 0.1103515625, | |
| "learning_rate": 0.00019075754196709572, | |
| "loss": 0.1436, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.47402725656725264, | |
| "grad_norm": 0.1025390625, | |
| "learning_rate": 0.0001904711706776006, | |
| "loss": 0.1408, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.48034761998814934, | |
| "grad_norm": 0.1171875, | |
| "learning_rate": 0.00019018065202237083, | |
| "loss": 0.1594, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.48666798340904605, | |
| "grad_norm": 0.10546875, | |
| "learning_rate": 0.00018988599931931866, | |
| "loss": 0.1394, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.49298834682994275, | |
| "grad_norm": 0.111328125, | |
| "learning_rate": 0.0001895872260758688, | |
| "loss": 0.1448, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.49930871025083945, | |
| "grad_norm": 0.111328125, | |
| "learning_rate": 0.00018928434598833912, | |
| "loss": 0.156, | |
| "step": 79 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 474, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 79, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.384934679033938e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |