| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.11355034065102196, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001514004542013626, | |
| "grad_norm": 3.2473833560943604, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1077, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001514004542013626, | |
| "eval_loss": 1.0687472820281982, | |
| "eval_runtime": 112.9034, | |
| "eval_samples_per_second": 4.925, | |
| "eval_steps_per_second": 0.62, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003028009084027252, | |
| "grad_norm": 3.2420403957366943, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0677, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.004542013626040878, | |
| "grad_norm": 3.1393091678619385, | |
| "learning_rate": 3e-05, | |
| "loss": 1.1821, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.006056018168054504, | |
| "grad_norm": 2.6600749492645264, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0311, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00757002271006813, | |
| "grad_norm": 3.0057172775268555, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0231, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.009084027252081756, | |
| "grad_norm": 2.4468352794647217, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9606, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.010598031794095382, | |
| "grad_norm": 1.6095988750457764, | |
| "learning_rate": 7e-05, | |
| "loss": 0.9656, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.012112036336109008, | |
| "grad_norm": 1.8507676124572754, | |
| "learning_rate": 8e-05, | |
| "loss": 0.9028, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.013626040878122634, | |
| "grad_norm": 1.5739439725875854, | |
| "learning_rate": 9e-05, | |
| "loss": 0.9131, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.013626040878122634, | |
| "eval_loss": 0.8630487322807312, | |
| "eval_runtime": 112.3821, | |
| "eval_samples_per_second": 4.947, | |
| "eval_steps_per_second": 0.623, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01514004542013626, | |
| "grad_norm": 1.68681800365448, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7878, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.016654049962149888, | |
| "grad_norm": 1.6374067068099976, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.8566, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.018168054504163512, | |
| "grad_norm": 1.1348960399627686, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.8245, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01968205904617714, | |
| "grad_norm": 0.9797458648681641, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.8307, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.021196063588190765, | |
| "grad_norm": 1.0176491737365723, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.8965, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.022710068130204392, | |
| "grad_norm": 1.0210918188095093, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.7902, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.024224072672218017, | |
| "grad_norm": 0.8656893968582153, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.7279, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.025738077214231644, | |
| "grad_norm": 0.888776421546936, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.877, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.02725208175624527, | |
| "grad_norm": 0.9443601369857788, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.8589, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02725208175624527, | |
| "eval_loss": 0.7156172394752502, | |
| "eval_runtime": 112.3775, | |
| "eval_samples_per_second": 4.948, | |
| "eval_steps_per_second": 0.623, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.028766086298258896, | |
| "grad_norm": 0.8223451972007751, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.7802, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.03028009084027252, | |
| "grad_norm": 1.0395427942276, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.7538, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03179409538228615, | |
| "grad_norm": 1.2328284978866577, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.6463, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.033308099924299776, | |
| "grad_norm": 0.8727930188179016, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.6779, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0348221044663134, | |
| "grad_norm": 0.7586942315101624, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.7326, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.036336109008327025, | |
| "grad_norm": 0.898556649684906, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.7108, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03785011355034065, | |
| "grad_norm": 0.7121559977531433, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.7086, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03936411809235428, | |
| "grad_norm": 0.9445357918739319, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.7351, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0408781226343679, | |
| "grad_norm": 0.7903055548667908, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.6486, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0408781226343679, | |
| "eval_loss": 0.6585749387741089, | |
| "eval_runtime": 112.3863, | |
| "eval_samples_per_second": 4.947, | |
| "eval_steps_per_second": 0.623, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.04239212717638153, | |
| "grad_norm": 0.7344522476196289, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.6864, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.04390613171839516, | |
| "grad_norm": 0.7938252091407776, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 0.5384, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.045420136260408785, | |
| "grad_norm": 0.8089536428451538, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.6791, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.046934140802422405, | |
| "grad_norm": 0.6889751553535461, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.6777, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.04844814534443603, | |
| "grad_norm": 0.9178895354270935, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.6124, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.04996214988644966, | |
| "grad_norm": 0.790649950504303, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.6654, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.05147615442846329, | |
| "grad_norm": 0.7298893928527832, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.6839, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.05299015897047691, | |
| "grad_norm": 0.7058828473091125, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.6835, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.05450416351249054, | |
| "grad_norm": 0.7319390773773193, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.6597, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.05450416351249054, | |
| "eval_loss": 0.6305856108665466, | |
| "eval_runtime": 112.421, | |
| "eval_samples_per_second": 4.946, | |
| "eval_steps_per_second": 0.623, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.056018168054504165, | |
| "grad_norm": 0.8053693175315857, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.7052, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.05753217259651779, | |
| "grad_norm": 0.8908604979515076, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.6315, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.059046177138531414, | |
| "grad_norm": 0.8547291159629822, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.55, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.06056018168054504, | |
| "grad_norm": 0.6951102614402771, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.6597, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.06207418622255867, | |
| "grad_norm": 0.8669103384017944, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.5783, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0635881907645723, | |
| "grad_norm": 0.7219836115837097, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.7272, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.06510219530658592, | |
| "grad_norm": 0.9448738694190979, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.5859, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.06661619984859955, | |
| "grad_norm": 0.8073548674583435, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.6732, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.06813020439061317, | |
| "grad_norm": 0.7021400332450867, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.5786, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06813020439061317, | |
| "eval_loss": 0.617196261882782, | |
| "eval_runtime": 112.3955, | |
| "eval_samples_per_second": 4.947, | |
| "eval_steps_per_second": 0.623, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0696442089326268, | |
| "grad_norm": 0.7012249827384949, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.7483, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.07115821347464042, | |
| "grad_norm": 0.7140128016471863, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.6484, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.07267221801665405, | |
| "grad_norm": 0.7511979341506958, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.5084, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.07418622255866768, | |
| "grad_norm": 0.9730421900749207, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.5932, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0757002271006813, | |
| "grad_norm": 0.7117515206336975, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.3311, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07721423164269493, | |
| "grad_norm": 0.6524476408958435, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.6501, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.07872823618470856, | |
| "grad_norm": 0.7036517262458801, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.6081, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.08024224072672217, | |
| "grad_norm": 0.6719091534614563, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 0.6072, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.0817562452687358, | |
| "grad_norm": 0.6507532000541687, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.6719, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0817562452687358, | |
| "eval_loss": 0.609633207321167, | |
| "eval_runtime": 112.3919, | |
| "eval_samples_per_second": 4.947, | |
| "eval_steps_per_second": 0.623, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.08327024981074943, | |
| "grad_norm": 0.755693256855011, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5745, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.08478425435276306, | |
| "grad_norm": 0.6789919137954712, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.5373, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.08629825889477669, | |
| "grad_norm": 0.6672856211662292, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.5594, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.08781226343679031, | |
| "grad_norm": 0.7143582105636597, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.581, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.08932626797880394, | |
| "grad_norm": 0.7472301721572876, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 0.5918, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.09084027252081757, | |
| "grad_norm": 0.6661199331283569, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.6445, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.09235427706283118, | |
| "grad_norm": 0.6965391635894775, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.6133, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.09386828160484481, | |
| "grad_norm": 0.8193009495735168, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.4107, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.09538228614685844, | |
| "grad_norm": 0.7017238140106201, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.6026, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.09538228614685844, | |
| "eval_loss": 0.6010457873344421, | |
| "eval_runtime": 112.3974, | |
| "eval_samples_per_second": 4.947, | |
| "eval_steps_per_second": 0.623, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.09689629068887207, | |
| "grad_norm": 0.8020389080047607, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.6733, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.0984102952308857, | |
| "grad_norm": 0.6975551247596741, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.5705, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.09992429977289932, | |
| "grad_norm": 0.719771146774292, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.6391, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.10143830431491295, | |
| "grad_norm": 0.7541303038597107, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 0.6722, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.10295230885692658, | |
| "grad_norm": 0.7610719203948975, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.6754, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.10446631339894019, | |
| "grad_norm": 0.7397011518478394, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.5281, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.10598031794095382, | |
| "grad_norm": 1.2769193649291992, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.6154, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.10749432248296745, | |
| "grad_norm": 0.6851471662521362, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 0.6214, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.10900832702498107, | |
| "grad_norm": 0.7720723152160645, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.6217, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.10900832702498107, | |
| "eval_loss": 0.59645015001297, | |
| "eval_runtime": 112.402, | |
| "eval_samples_per_second": 4.947, | |
| "eval_steps_per_second": 0.623, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1105223315669947, | |
| "grad_norm": 0.7584350109100342, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.6332, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.11203633610900833, | |
| "grad_norm": 0.6037428975105286, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.6377, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.11355034065102196, | |
| "grad_norm": 0.6927869915962219, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.6652, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0275138877312205e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |