| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.3624022305011749, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5345, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.3736283481121063, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5545, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.3715187907218933, | |
| "learning_rate": 6e-05, | |
| "loss": 0.5167, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.3104977607727051, | |
| "learning_rate": 8e-05, | |
| "loss": 0.4994, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.3549410104751587, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5208, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.41164880990982056, | |
| "learning_rate": 9.997266286704631e-05, | |
| "loss": 0.528, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.4071100354194641, | |
| "learning_rate": 9.989068136093873e-05, | |
| "loss": 0.5083, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.42801719903945923, | |
| "learning_rate": 9.975414512725057e-05, | |
| "loss": 0.5336, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.43720319867134094, | |
| "learning_rate": 9.956320346634876e-05, | |
| "loss": 0.5478, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.39605292677879333, | |
| "learning_rate": 9.931806517013612e-05, | |
| "loss": 0.5509, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.4373878240585327, | |
| "learning_rate": 9.901899829374047e-05, | |
| "loss": 0.4953, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.3865221440792084, | |
| "learning_rate": 9.86663298624003e-05, | |
| "loss": 0.5273, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.3498033881187439, | |
| "learning_rate": 9.826044551386744e-05, | |
| "loss": 0.5626, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.33049601316452026, | |
| "learning_rate": 9.780178907671789e-05, | |
| "loss": 0.5252, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.3525201082229614, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 0.5433, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.3323298990726471, | |
| "learning_rate": 9.672822322997305e-05, | |
| "loss": 0.5371, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.34475523233413696, | |
| "learning_rate": 9.611448774886924e-05, | |
| "loss": 0.5273, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.33978456258773804, | |
| "learning_rate": 9.545032675245813e-05, | |
| "loss": 0.5377, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.32446014881134033, | |
| "learning_rate": 9.473646649103818e-05, | |
| "loss": 0.5321, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.32642048597335815, | |
| "learning_rate": 9.397368756032445e-05, | |
| "loss": 0.5134, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.32789453864097595, | |
| "learning_rate": 9.316282404787871e-05, | |
| "loss": 0.5516, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.33384475111961365, | |
| "learning_rate": 9.230476262104677e-05, | |
| "loss": 0.5656, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.3290055990219116, | |
| "learning_rate": 9.140044155740101e-05, | |
| "loss": 0.5809, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.3402266204357147, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.5617, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.33358535170555115, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 0.5357, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.33832627534866333, | |
| "learning_rate": 8.842005554284296e-05, | |
| "loss": 0.5587, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.3213883638381958, | |
| "learning_rate": 8.73410738492077e-05, | |
| "loss": 0.5697, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.34648266434669495, | |
| "learning_rate": 8.622126023955446e-05, | |
| "loss": 0.522, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.3362743854522705, | |
| "learning_rate": 8.506183921362443e-05, | |
| "loss": 0.5572, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.3271280825138092, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 0.514, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.334176629781723, | |
| "learning_rate": 8.262928807620843e-05, | |
| "loss": 0.5092, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.33813613653182983, | |
| "learning_rate": 8.135881792367686e-05, | |
| "loss": 0.57, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.3526760935783386, | |
| "learning_rate": 8.005405736415126e-05, | |
| "loss": 0.5737, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.3210213780403137, | |
| "learning_rate": 7.871643313414718e-05, | |
| "loss": 0.5444, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.3287356197834015, | |
| "learning_rate": 7.734740790612136e-05, | |
| "loss": 0.5188, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.3406030535697937, | |
| "learning_rate": 7.594847868906076e-05, | |
| "loss": 0.5491, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.3229946792125702, | |
| "learning_rate": 7.452117519152542e-05, | |
| "loss": 0.5309, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.31070253252983093, | |
| "learning_rate": 7.30670581489344e-05, | |
| "loss": 0.5348, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.31888288259506226, | |
| "learning_rate": 7.158771761692464e-05, | |
| "loss": 0.5598, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.3222619295120239, | |
| "learning_rate": 7.008477123264848e-05, | |
| "loss": 0.5674, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.3319852352142334, | |
| "learning_rate": 6.855986244591104e-05, | |
| "loss": 0.5371, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.30418190360069275, | |
| "learning_rate": 6.701465872208216e-05, | |
| "loss": 0.5104, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.3073510229587555, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.5359, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.3012188673019409, | |
| "learning_rate": 6.387014543809223e-05, | |
| "loss": 0.5341, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.2977622151374817, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 0.5448, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.312294065952301, | |
| "learning_rate": 6.066498153718735e-05, | |
| "loss": 0.5421, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.33205491304397583, | |
| "learning_rate": 5.90440267166055e-05, | |
| "loss": 0.5841, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.31330958008766174, | |
| "learning_rate": 5.74131823855921e-05, | |
| "loss": 0.5344, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.32422611117362976, | |
| "learning_rate": 5.577423184847932e-05, | |
| "loss": 0.5277, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.3156668245792389, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 0.5454, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 0.3290346562862396, | |
| "learning_rate": 5.247918773366112e-05, | |
| "loss": 0.4873, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.3424556255340576, | |
| "learning_rate": 5.0826697238317935e-05, | |
| "loss": 0.4639, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 0.31981292366981506, | |
| "learning_rate": 4.917330276168208e-05, | |
| "loss": 0.459, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.3340217173099518, | |
| "learning_rate": 4.7520812266338885e-05, | |
| "loss": 0.4981, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.3345750868320465, | |
| "learning_rate": 4.5871032726383386e-05, | |
| "loss": 0.4898, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.2998408079147339, | |
| "learning_rate": 4.4225768151520694e-05, | |
| "loss": 0.4329, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.1400000000000001, | |
| "grad_norm": 0.33386602997779846, | |
| "learning_rate": 4.2586817614407895e-05, | |
| "loss": 0.43, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.3249485492706299, | |
| "learning_rate": 4.095597328339452e-05, | |
| "loss": 0.4762, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 0.32801133394241333, | |
| "learning_rate": 3.933501846281267e-05, | |
| "loss": 0.4514, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.32089194655418396, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 0.4206, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 0.32813841104507446, | |
| "learning_rate": 3.612985456190778e-05, | |
| "loss": 0.4853, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.33598625659942627, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.4334, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 0.3225085139274597, | |
| "learning_rate": 3.298534127791785e-05, | |
| "loss": 0.4479, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.3295815885066986, | |
| "learning_rate": 3.144013755408895e-05, | |
| "loss": 0.4739, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.3391561508178711, | |
| "learning_rate": 2.991522876735154e-05, | |
| "loss": 0.4211, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 0.3367979824542999, | |
| "learning_rate": 2.8412282383075363e-05, | |
| "loss": 0.4502, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 0.3409813344478607, | |
| "learning_rate": 2.693294185106562e-05, | |
| "loss": 0.4886, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 0.33046984672546387, | |
| "learning_rate": 2.547882480847461e-05, | |
| "loss": 0.4142, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 0.3314824104309082, | |
| "learning_rate": 2.405152131093926e-05, | |
| "loss": 0.4535, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.34188470244407654, | |
| "learning_rate": 2.2652592093878666e-05, | |
| "loss": 0.4525, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 0.3405551612377167, | |
| "learning_rate": 2.128356686585282e-05, | |
| "loss": 0.4306, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.34439051151275635, | |
| "learning_rate": 1.9945942635848748e-05, | |
| "loss": 0.4849, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 0.33053305745124817, | |
| "learning_rate": 1.8641182076323148e-05, | |
| "loss": 0.4395, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.3231377601623535, | |
| "learning_rate": 1.7370711923791567e-05, | |
| "loss": 0.4461, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.3283744752407074, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 0.4322, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.3327557146549225, | |
| "learning_rate": 1.4938160786375572e-05, | |
| "loss": 0.4796, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 0.3517839312553406, | |
| "learning_rate": 1.3778739760445552e-05, | |
| "loss": 0.4617, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.3397415578365326, | |
| "learning_rate": 1.2658926150792322e-05, | |
| "loss": 0.4561, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 0.32547858357429504, | |
| "learning_rate": 1.157994445715706e-05, | |
| "loss": 0.4522, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.322001576423645, | |
| "learning_rate": 1.0542974530180327e-05, | |
| "loss": 0.4275, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 0.32253459095954895, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.4388, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.33204764127731323, | |
| "learning_rate": 8.599558442598998e-06, | |
| "loss": 0.4317, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.6600000000000001, | |
| "grad_norm": 0.32604044675827026, | |
| "learning_rate": 7.695237378953223e-06, | |
| "loss": 0.4357, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 0.34298455715179443, | |
| "learning_rate": 6.837175952121306e-06, | |
| "loss": 0.442, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 0.3202356994152069, | |
| "learning_rate": 6.026312439675552e-06, | |
| "loss": 0.4342, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.3258272409439087, | |
| "learning_rate": 5.263533508961827e-06, | |
| "loss": 0.4348, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 0.3249031901359558, | |
| "learning_rate": 4.549673247541875e-06, | |
| "loss": 0.4093, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.32874855399131775, | |
| "learning_rate": 3.885512251130763e-06, | |
| "loss": 0.4076, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 0.33212965726852417, | |
| "learning_rate": 3.271776770026963e-06, | |
| "loss": 0.4426, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.32816311717033386, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 0.4422, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.8199999999999998, | |
| "grad_norm": 0.333683580160141, | |
| "learning_rate": 2.1982109232821178e-06, | |
| "loss": 0.4619, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 0.3273014724254608, | |
| "learning_rate": 1.7395544861325718e-06, | |
| "loss": 0.4662, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.8599999999999999, | |
| "grad_norm": 0.33303171396255493, | |
| "learning_rate": 1.333670137599713e-06, | |
| "loss": 0.4467, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.3208991587162018, | |
| "learning_rate": 9.810017062595322e-07, | |
| "loss": 0.4459, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.3281053304672241, | |
| "learning_rate": 6.819348298638839e-07, | |
| "loss": 0.4471, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.31943386793136597, | |
| "learning_rate": 4.367965336512403e-07, | |
| "loss": 0.4342, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 0.3442579507827759, | |
| "learning_rate": 2.458548727494292e-07, | |
| "loss": 0.454, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.3283260762691498, | |
| "learning_rate": 1.0931863906127327e-07, | |
| "loss": 0.4328, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 0.33244219422340393, | |
| "learning_rate": 2.7337132953697554e-08, | |
| "loss": 0.4145, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.3326053321361542, | |
| "learning_rate": 0.0, | |
| "loss": 0.4836, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "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": 1.211706059414569e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |