| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 90, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.022222222222222223, | |
| "grad_norm": 0.15352968871988337, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1828, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.044444444444444446, | |
| "grad_norm": 0.1992673485223412, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.1999, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.06666666666666667, | |
| "grad_norm": 0.1672953053939959, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.1535, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08888888888888889, | |
| "grad_norm": 0.27881675369340553, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.2579, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 0.19015211653657962, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1451, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.13333333333333333, | |
| "grad_norm": 0.17685374617481667, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1233, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.15555555555555556, | |
| "grad_norm": 0.11511643797638078, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.0668, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.17777777777777778, | |
| "grad_norm": 0.35336946985881396, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.3579, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.36377699724925733, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3149, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 0.13466476239074193, | |
| "learning_rate": 4.998119881260576e-05, | |
| "loss": 0.0576, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.24444444444444444, | |
| "grad_norm": 0.1927678314689982, | |
| "learning_rate": 4.99248235291948e-05, | |
| "loss": 0.1805, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "grad_norm": 0.12324214014637602, | |
| "learning_rate": 4.983095894354858e-05, | |
| "loss": 0.081, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.28888888888888886, | |
| "grad_norm": 0.3239776824825179, | |
| "learning_rate": 4.969974623692023e-05, | |
| "loss": 0.26, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.3111111111111111, | |
| "grad_norm": 0.2155401569773017, | |
| "learning_rate": 4.953138276568462e-05, | |
| "loss": 0.1257, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 0.2897815766964899, | |
| "learning_rate": 4.9326121764495596e-05, | |
| "loss": 0.0976, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 0.39829676265862246, | |
| "learning_rate": 4.9084271965397014e-05, | |
| "loss": 0.143, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.37777777777777777, | |
| "grad_norm": 0.20210191072903955, | |
| "learning_rate": 4.880619713346039e-05, | |
| "loss": 0.1066, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.40880170866877663, | |
| "learning_rate": 4.849231551964771e-05, | |
| "loss": 0.2189, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.4222222222222222, | |
| "grad_norm": 0.4184595635898202, | |
| "learning_rate": 4.814309923172227e-05, | |
| "loss": 0.1845, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 0.2715390485387202, | |
| "learning_rate": 4.775907352415367e-05, | |
| "loss": 0.1263, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4666666666666667, | |
| "grad_norm": 0.19016690002806513, | |
| "learning_rate": 4.734081600808531e-05, | |
| "loss": 0.0394, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.4888888888888889, | |
| "grad_norm": 0.3806812284355486, | |
| "learning_rate": 4.6888955782552274e-05, | |
| "loss": 0.2056, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.5111111111111111, | |
| "grad_norm": 0.26875087498837735, | |
| "learning_rate": 4.640417248825667e-05, | |
| "loss": 0.1077, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "grad_norm": 0.22530421732750502, | |
| "learning_rate": 4.588719528532342e-05, | |
| "loss": 0.1043, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 0.18405476029818285, | |
| "learning_rate": 4.533880175657419e-05, | |
| "loss": 0.0795, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.5777777777777777, | |
| "grad_norm": 0.25568790339476477, | |
| "learning_rate": 4.475981673796899e-05, | |
| "loss": 0.1235, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.1977921472382721, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.0909, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.6222222222222222, | |
| "grad_norm": 0.2605021359404801, | |
| "learning_rate": 4.351360032772512e-05, | |
| "loss": 0.0844, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.6444444444444445, | |
| "grad_norm": 0.1363348491849494, | |
| "learning_rate": 4.2848243363947484e-05, | |
| "loss": 0.0419, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 0.25954324302915655, | |
| "learning_rate": 4.215604094671835e-05, | |
| "loss": 0.083, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6888888888888889, | |
| "grad_norm": 0.15241320837389336, | |
| "learning_rate": 4.14380342142266e-05, | |
| "loss": 0.0505, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.21196345448497633, | |
| "learning_rate": 4.069530311680247e-05, | |
| "loss": 0.0458, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.7333333333333333, | |
| "grad_norm": 0.14671622354492578, | |
| "learning_rate": 3.9928964792569655e-05, | |
| "loss": 0.0515, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.7555555555555555, | |
| "grad_norm": 0.24886874701283312, | |
| "learning_rate": 3.914017188716347e-05, | |
| "loss": 0.0779, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.7777777777777778, | |
| "grad_norm": 0.2759509508684298, | |
| "learning_rate": 3.8330110820042285e-05, | |
| "loss": 0.064, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.1455636144453065, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.07, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.8222222222222222, | |
| "grad_norm": 0.2708881899406701, | |
| "learning_rate": 3.665108799256348e-05, | |
| "loss": 0.1154, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.8444444444444444, | |
| "grad_norm": 0.20701892065585925, | |
| "learning_rate": 3.578465164203134e-05, | |
| "loss": 0.064, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.8666666666666667, | |
| "grad_norm": 0.07525579966677097, | |
| "learning_rate": 3.490199415097892e-05, | |
| "loss": 0.0174, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 0.47775650060704056, | |
| "learning_rate": 3.400444312011776e-05, | |
| "loss": 0.2248, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.9111111111111111, | |
| "grad_norm": 0.1512821048275736, | |
| "learning_rate": 3.309334855145803e-05, | |
| "loss": 0.048, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.9333333333333333, | |
| "grad_norm": 0.2668654620386901, | |
| "learning_rate": 3.217008081777726e-05, | |
| "loss": 0.0612, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.9555555555555556, | |
| "grad_norm": 0.20792461168152554, | |
| "learning_rate": 3.1236028601449534e-05, | |
| "loss": 0.0744, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.9777777777777777, | |
| "grad_norm": 0.1471973464840355, | |
| "learning_rate": 3.0292596805735274e-05, | |
| "loss": 0.0323, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.15571330433687014, | |
| "learning_rate": 2.9341204441673266e-05, | |
| "loss": 0.0395, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.13151995837688446, | |
| "eval_runtime": 0.3971, | |
| "eval_samples_per_second": 2.519, | |
| "eval_steps_per_second": 2.519, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0222222222222221, | |
| "grad_norm": 0.08463452690558734, | |
| "learning_rate": 2.8383282493753283e-05, | |
| "loss": 0.0204, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.0444444444444445, | |
| "grad_norm": 0.12410620903011262, | |
| "learning_rate": 2.742027176757948e-05, | |
| "loss": 0.0311, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.057661254660238734, | |
| "learning_rate": 2.6453620722761896e-05, | |
| "loss": 0.0117, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.0888888888888888, | |
| "grad_norm": 0.11691070820752421, | |
| "learning_rate": 2.548478329429561e-05, | |
| "loss": 0.0384, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 0.19002974743608797, | |
| "learning_rate": 2.4515216705704395e-05, | |
| "loss": 0.0533, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1333333333333333, | |
| "grad_norm": 0.11112574882832801, | |
| "learning_rate": 2.3546379277238107e-05, | |
| "loss": 0.0228, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.1555555555555554, | |
| "grad_norm": 0.18194491846874342, | |
| "learning_rate": 2.2579728232420525e-05, | |
| "loss": 0.0583, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.1777777777777778, | |
| "grad_norm": 0.1533461648182278, | |
| "learning_rate": 2.161671750624673e-05, | |
| "loss": 0.0465, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.06909879137605512, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.011, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.2222222222222223, | |
| "grad_norm": 0.04652100573581006, | |
| "learning_rate": 1.970740319426474e-05, | |
| "loss": 0.0082, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.2444444444444445, | |
| "grad_norm": 0.07851036278058815, | |
| "learning_rate": 1.876397139855047e-05, | |
| "loss": 0.0136, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.2666666666666666, | |
| "grad_norm": 0.3387532671795071, | |
| "learning_rate": 1.7829919182222752e-05, | |
| "loss": 0.1106, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.2888888888888888, | |
| "grad_norm": 0.14394041528242452, | |
| "learning_rate": 1.690665144854198e-05, | |
| "loss": 0.0259, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.3111111111111111, | |
| "grad_norm": 0.12193751882013566, | |
| "learning_rate": 1.5995556879882246e-05, | |
| "loss": 0.0357, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 0.10245566375207657, | |
| "learning_rate": 1.509800584902108e-05, | |
| "loss": 0.0294, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.3555555555555556, | |
| "grad_norm": 0.1549289530175864, | |
| "learning_rate": 1.4215348357968669e-05, | |
| "loss": 0.0301, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.3777777777777778, | |
| "grad_norm": 0.0510702557495791, | |
| "learning_rate": 1.3348912007436537e-05, | |
| "loss": 0.0053, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.09129828566612887, | |
| "learning_rate": 1.2500000000000006e-05, | |
| "loss": 0.0095, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.11685635904352058, | |
| "learning_rate": 1.1669889179957725e-05, | |
| "loss": 0.0263, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.4444444444444444, | |
| "grad_norm": 0.059282074482053033, | |
| "learning_rate": 1.085982811283654e-05, | |
| "loss": 0.0126, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.4666666666666668, | |
| "grad_norm": 0.04492683333176809, | |
| "learning_rate": 1.0071035207430352e-05, | |
| "loss": 0.0063, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.488888888888889, | |
| "grad_norm": 0.07991980856274734, | |
| "learning_rate": 9.304696883197542e-06, | |
| "loss": 0.0168, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.511111111111111, | |
| "grad_norm": 0.18860273052381646, | |
| "learning_rate": 8.561965785773413e-06, | |
| "loss": 0.0138, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.5333333333333332, | |
| "grad_norm": 0.04637738493485198, | |
| "learning_rate": 7.843959053281663e-06, | |
| "loss": 0.0059, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.5555555555555556, | |
| "grad_norm": 0.11241673386744654, | |
| "learning_rate": 7.1517566360525284e-06, | |
| "loss": 0.0167, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.5777777777777777, | |
| "grad_norm": 0.06513269256685703, | |
| "learning_rate": 6.48639967227489e-06, | |
| "loss": 0.0121, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.1499902093509071, | |
| "learning_rate": 5.848888922025553e-06, | |
| "loss": 0.029, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.6222222222222222, | |
| "grad_norm": 0.025048176137643145, | |
| "learning_rate": 5.240183262031021e-06, | |
| "loss": 0.004, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.6444444444444444, | |
| "grad_norm": 0.1211626973278271, | |
| "learning_rate": 4.661198243425813e-06, | |
| "loss": 0.0156, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.3541302340517843, | |
| "learning_rate": 4.112804714676594e-06, | |
| "loss": 0.0247, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.6888888888888889, | |
| "grad_norm": 0.10275595190598034, | |
| "learning_rate": 3.595827511743341e-06, | |
| "loss": 0.0149, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.7111111111111112, | |
| "grad_norm": 0.021989766789088025, | |
| "learning_rate": 3.111044217447731e-06, | |
| "loss": 0.0031, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.7333333333333334, | |
| "grad_norm": 0.05086015561609853, | |
| "learning_rate": 2.659183991914696e-06, | |
| "loss": 0.0082, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.7555555555555555, | |
| "grad_norm": 0.1233758707928277, | |
| "learning_rate": 2.2409264758463363e-06, | |
| "loss": 0.0163, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.12858923331151956, | |
| "learning_rate": 1.8569007682777417e-06, | |
| "loss": 0.0231, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.10088342416382587, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.0117, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.8222222222222222, | |
| "grad_norm": 0.10722303391126128, | |
| "learning_rate": 1.1938028665396173e-06, | |
| "loss": 0.0247, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.8444444444444446, | |
| "grad_norm": 0.1830696778117431, | |
| "learning_rate": 9.157280346029918e-07, | |
| "loss": 0.0264, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.8666666666666667, | |
| "grad_norm": 0.1259648620962399, | |
| "learning_rate": 6.738782355044049e-07, | |
| "loss": 0.032, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.8888888888888888, | |
| "grad_norm": 0.11980211541006067, | |
| "learning_rate": 4.6861723431538276e-07, | |
| "loss": 0.0322, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.911111111111111, | |
| "grad_norm": 0.23359207984719352, | |
| "learning_rate": 3.002537630797747e-07, | |
| "loss": 0.0425, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.9333333333333333, | |
| "grad_norm": 0.10948078113683882, | |
| "learning_rate": 1.6904105645142444e-07, | |
| "loss": 0.019, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.9555555555555557, | |
| "grad_norm": 0.040764819228826535, | |
| "learning_rate": 7.51764708051994e-08, | |
| "loss": 0.0064, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.9777777777777779, | |
| "grad_norm": 0.015632399712207767, | |
| "learning_rate": 1.8801187394248965e-08, | |
| "loss": 0.0021, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.10746360158455943, | |
| "learning_rate": 0.0, | |
| "loss": 0.0115, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.15711571276187897, | |
| "eval_runtime": 1.4509, | |
| "eval_samples_per_second": 0.689, | |
| "eval_steps_per_second": 0.689, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 90, | |
| "total_flos": 422804029440.0, | |
| "train_loss": 0.07111663229815247, | |
| "train_runtime": 202.1032, | |
| "train_samples_per_second": 0.891, | |
| "train_steps_per_second": 0.445 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 90, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 422804029440.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |