| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.736842105263158, | |
| "eval_steps": 500, | |
| "global_step": 81, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.10526315789473684, | |
| "grad_norm": 14.31118392944336, | |
| "learning_rate": 2e-05, | |
| "loss": 2.4434, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "grad_norm": 10.385307312011719, | |
| "learning_rate": 4e-05, | |
| "loss": 2.5855, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "grad_norm": 18.257061004638672, | |
| "learning_rate": 6e-05, | |
| "loss": 2.5327, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.42105263157894735, | |
| "grad_norm": 9.803607940673828, | |
| "learning_rate": 8e-05, | |
| "loss": 2.401, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5263157894736842, | |
| "grad_norm": 10.1410493850708, | |
| "learning_rate": 0.0001, | |
| "loss": 2.3054, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "grad_norm": 8.388815879821777, | |
| "learning_rate": 0.00012, | |
| "loss": 2.3298, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7368421052631579, | |
| "grad_norm": 7.938385963439941, | |
| "learning_rate": 0.00014, | |
| "loss": 2.0323, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "grad_norm": 7.110404968261719, | |
| "learning_rate": 0.00016, | |
| "loss": 2.0579, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "grad_norm": 3.487077474594116, | |
| "learning_rate": 0.00018, | |
| "loss": 1.6501, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.0789473684210527, | |
| "grad_norm": 5.468748092651367, | |
| "learning_rate": 0.0002, | |
| "loss": 2.8277, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.1842105263157894, | |
| "grad_norm": 2.3874928951263428, | |
| "learning_rate": 0.0001999229036240723, | |
| "loss": 1.2231, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.2894736842105263, | |
| "grad_norm": 5.447896480560303, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 1.5471, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.3947368421052633, | |
| "grad_norm": 3.8631601333618164, | |
| "learning_rate": 0.00019930684569549264, | |
| "loss": 1.1826, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 12.651008605957031, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 1.4406, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.6052631578947367, | |
| "grad_norm": 2.3405802249908447, | |
| "learning_rate": 0.00019807852804032305, | |
| "loss": 1.0181, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.7105263157894737, | |
| "grad_norm": 6.5888671875, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 1.2595, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.8157894736842106, | |
| "grad_norm": 4.1633758544921875, | |
| "learning_rate": 0.00019624552364536473, | |
| "loss": 1.0053, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.9210526315789473, | |
| "grad_norm": 2.3248441219329834, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.9642, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.0526315789473686, | |
| "grad_norm": 5.530248641967773, | |
| "learning_rate": 0.00019381913359224842, | |
| "loss": 2.2521, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.1578947368421053, | |
| "grad_norm": 1.3485472202301025, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.7636, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.263157894736842, | |
| "grad_norm": 2.785285472869873, | |
| "learning_rate": 0.00019081431738250814, | |
| "loss": 1.0499, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.3684210526315788, | |
| "grad_norm": 2.2571582794189453, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.8719, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.473684210526316, | |
| "grad_norm": 1.8401918411254883, | |
| "learning_rate": 0.00018724960070727972, | |
| "loss": 0.8511, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.5789473684210527, | |
| "grad_norm": 2.386539936065674, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.8703, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.6842105263157894, | |
| "grad_norm": 1.384438157081604, | |
| "learning_rate": 0.00018314696123025454, | |
| "loss": 0.6987, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.7894736842105265, | |
| "grad_norm": 1.705210566520691, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.8401, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.8947368421052633, | |
| "grad_norm": 1.8555241823196411, | |
| "learning_rate": 0.00017853169308807448, | |
| "loss": 1.023, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 3.026315789473684, | |
| "grad_norm": 3.781007766723633, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 1.4192, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.1315789473684212, | |
| "grad_norm": 1.589735507965088, | |
| "learning_rate": 0.00017343225094356855, | |
| "loss": 0.7493, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 3.236842105263158, | |
| "grad_norm": 2.163848638534546, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 1.0892, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.3421052631578947, | |
| "grad_norm": 1.8226677179336548, | |
| "learning_rate": 0.0001678800745532942, | |
| "loss": 0.7154, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 3.4473684210526314, | |
| "grad_norm": 1.5850944519042969, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.8619, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.5526315789473686, | |
| "grad_norm": 1.5230872631072998, | |
| "learning_rate": 0.00016190939493098344, | |
| "loss": 0.6566, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.6578947368421053, | |
| "grad_norm": 1.870915174484253, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.8506, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.763157894736842, | |
| "grad_norm": 1.698755145072937, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 0.762, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.8684210526315788, | |
| "grad_norm": 1.6367840766906738, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.7136, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.973684210526316, | |
| "grad_norm": 2.9744303226470947, | |
| "learning_rate": 0.00014886212414969553, | |
| "loss": 0.9568, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 4.105263157894737, | |
| "grad_norm": 1.3073623180389404, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.6397, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.2105263157894735, | |
| "grad_norm": 1.2846301794052124, | |
| "learning_rate": 0.0001418659737537428, | |
| "loss": 0.4418, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 4.315789473684211, | |
| "grad_norm": 1.6083595752716064, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.7064, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.421052631578947, | |
| "grad_norm": 1.2033262252807617, | |
| "learning_rate": 0.0001346117057077493, | |
| "loss": 0.5957, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 4.526315789473684, | |
| "grad_norm": 1.3466635942459106, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.7566, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.631578947368421, | |
| "grad_norm": 1.616913914680481, | |
| "learning_rate": 0.00012714404498650743, | |
| "loss": 0.7539, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 4.7368421052631575, | |
| "grad_norm": 1.4130221605300903, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.4622, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.842105263157895, | |
| "grad_norm": 1.3309547901153564, | |
| "learning_rate": 0.00011950903220161285, | |
| "loss": 0.7263, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 4.947368421052632, | |
| "grad_norm": 1.354387879371643, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.5625, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.078947368421052, | |
| "grad_norm": 3.008500337600708, | |
| "learning_rate": 0.00011175373974578378, | |
| "loss": 0.7619, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 5.184210526315789, | |
| "grad_norm": 1.5682655572891235, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.3858, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 5.2894736842105265, | |
| "grad_norm": 1.8495001792907715, | |
| "learning_rate": 0.00010392598157590688, | |
| "loss": 0.4573, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 5.394736842105263, | |
| "grad_norm": 2.635056972503662, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7921, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "grad_norm": 1.5783675909042358, | |
| "learning_rate": 9.607401842409317e-05, | |
| "loss": 0.3204, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 5.605263157894737, | |
| "grad_norm": 2.0661301612854004, | |
| "learning_rate": 9.215409042721552e-05, | |
| "loss": 0.6708, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 5.7105263157894735, | |
| "grad_norm": 1.4969799518585205, | |
| "learning_rate": 8.824626025421626e-05, | |
| "loss": 0.4052, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 5.815789473684211, | |
| "grad_norm": 1.6124238967895508, | |
| "learning_rate": 8.435655349597689e-05, | |
| "loss": 0.4829, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.921052631578947, | |
| "grad_norm": 1.431242823600769, | |
| "learning_rate": 8.049096779838719e-05, | |
| "loss": 0.4273, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 6.052631578947368, | |
| "grad_norm": 3.2095937728881836, | |
| "learning_rate": 7.66554636144095e-05, | |
| "loss": 0.8437, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 6.157894736842105, | |
| "grad_norm": 0.8892129063606262, | |
| "learning_rate": 7.285595501349258e-05, | |
| "loss": 0.2127, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 6.2631578947368425, | |
| "grad_norm": 1.6543368101119995, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.4305, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 6.368421052631579, | |
| "grad_norm": 1.4523223638534546, | |
| "learning_rate": 6.538829429225069e-05, | |
| "loss": 0.3762, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 6.473684210526316, | |
| "grad_norm": 0.9503556489944458, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.3628, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.578947368421053, | |
| "grad_norm": 1.4268410205841064, | |
| "learning_rate": 5.8134026246257225e-05, | |
| "loss": 0.4751, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 6.684210526315789, | |
| "grad_norm": 0.9671533703804016, | |
| "learning_rate": 5.4600950026045326e-05, | |
| "loss": 0.2934, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 6.7894736842105265, | |
| "grad_norm": 1.1906907558441162, | |
| "learning_rate": 5.113787585030454e-05, | |
| "loss": 0.2345, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 6.894736842105263, | |
| "grad_norm": 1.5449588298797607, | |
| "learning_rate": 4.7750143528405126e-05, | |
| "loss": 0.4283, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 7.026315789473684, | |
| "grad_norm": 2.223735809326172, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 0.4928, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 7.131578947368421, | |
| "grad_norm": 0.9985809326171875, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.3185, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 7.2368421052631575, | |
| "grad_norm": 1.0797033309936523, | |
| "learning_rate": 3.8090605069016595e-05, | |
| "loss": 0.4603, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 7.342105263157895, | |
| "grad_norm": 1.0436688661575317, | |
| "learning_rate": 3.5055195166981645e-05, | |
| "loss": 0.4009, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 7.447368421052632, | |
| "grad_norm": 1.1368814706802368, | |
| "learning_rate": 3.211992544670582e-05, | |
| "loss": 0.2486, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 7.552631578947368, | |
| "grad_norm": 1.0792099237442017, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.3308, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 7.657894736842105, | |
| "grad_norm": 0.9220963716506958, | |
| "learning_rate": 2.6567749056431467e-05, | |
| "loss": 0.3044, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 7.7631578947368425, | |
| "grad_norm": 1.1355466842651367, | |
| "learning_rate": 2.3959403439996907e-05, | |
| "loss": 0.2657, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 7.868421052631579, | |
| "grad_norm": 0.9300779104232788, | |
| "learning_rate": 2.146830691192553e-05, | |
| "loss": 0.2587, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 7.973684210526316, | |
| "grad_norm": 2.0543935298919678, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.5869, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 8.105263157894736, | |
| "grad_norm": 1.0328178405761719, | |
| "learning_rate": 1.6853038769745467e-05, | |
| "loss": 0.3424, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 8.210526315789474, | |
| "grad_norm": 1.0409759283065796, | |
| "learning_rate": 1.4735983564590783e-05, | |
| "loss": 0.3577, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 8.31578947368421, | |
| "grad_norm": 1.2581232786178589, | |
| "learning_rate": 1.2750399292720283e-05, | |
| "loss": 0.354, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 8.421052631578947, | |
| "grad_norm": 1.073712706565857, | |
| "learning_rate": 1.0899347581163221e-05, | |
| "loss": 0.3332, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 8.526315789473685, | |
| "grad_norm": 1.0400298833847046, | |
| "learning_rate": 9.185682617491863e-06, | |
| "loss": 0.3112, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 8.631578947368421, | |
| "grad_norm": 1.7903478145599365, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.4138, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 8.736842105263158, | |
| "grad_norm": 1.1479917764663696, | |
| "learning_rate": 6.180866407751595e-06, | |
| "loss": 0.2799, | |
| "step": 81 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 90, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0747062240097075e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |