| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.75609756097561, | |
| "eval_steps": 5, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04878048780487805, | |
| "grad_norm": 5.151791580979949, | |
| "learning_rate": 5e-08, | |
| "loss": 2.9815, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.24390243902439024, | |
| "grad_norm": 5.13763305558627, | |
| "learning_rate": 2.5e-07, | |
| "loss": 2.9732, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.24390243902439024, | |
| "eval_loss": 2.9549503326416016, | |
| "eval_runtime": 23.256, | |
| "eval_samples_per_second": 31.734, | |
| "eval_steps_per_second": 0.817, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.4878048780487805, | |
| "grad_norm": 4.3040377415100535, | |
| "learning_rate": 5e-07, | |
| "loss": 2.9685, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4878048780487805, | |
| "eval_loss": 2.9328665733337402, | |
| "eval_runtime": 19.8689, | |
| "eval_samples_per_second": 37.144, | |
| "eval_steps_per_second": 0.956, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7317073170731707, | |
| "grad_norm": 3.403638208876821, | |
| "learning_rate": 7.5e-07, | |
| "loss": 2.9341, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.7317073170731707, | |
| "eval_loss": 2.886631488800049, | |
| "eval_runtime": 18.4607, | |
| "eval_samples_per_second": 39.977, | |
| "eval_steps_per_second": 1.029, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.975609756097561, | |
| "grad_norm": 3.00111206949746, | |
| "learning_rate": 1e-06, | |
| "loss": 2.8788, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.975609756097561, | |
| "eval_loss": 2.8079330921173096, | |
| "eval_runtime": 19.9381, | |
| "eval_samples_per_second": 37.014, | |
| "eval_steps_per_second": 0.953, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "grad_norm": 2.7591628779527766, | |
| "learning_rate": 9.980973490458728e-07, | |
| "loss": 2.8082, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "eval_loss": 2.7484195232391357, | |
| "eval_runtime": 18.8326, | |
| "eval_samples_per_second": 39.187, | |
| "eval_steps_per_second": 1.009, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.4634146341463414, | |
| "grad_norm": 2.388152230770368, | |
| "learning_rate": 9.92403876506104e-07, | |
| "loss": 2.7341, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.4634146341463414, | |
| "eval_loss": 2.6838204860687256, | |
| "eval_runtime": 20.3367, | |
| "eval_samples_per_second": 36.289, | |
| "eval_steps_per_second": 0.934, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.7073170731707317, | |
| "grad_norm": 2.339030304927524, | |
| "learning_rate": 9.82962913144534e-07, | |
| "loss": 2.6784, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.7073170731707317, | |
| "eval_loss": 2.633502244949341, | |
| "eval_runtime": 18.7828, | |
| "eval_samples_per_second": 39.291, | |
| "eval_steps_per_second": 1.012, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.951219512195122, | |
| "grad_norm": 2.289213281990378, | |
| "learning_rate": 9.698463103929541e-07, | |
| "loss": 2.6326, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.951219512195122, | |
| "eval_loss": 2.5951168537139893, | |
| "eval_runtime": 20.4763, | |
| "eval_samples_per_second": 36.042, | |
| "eval_steps_per_second": 0.928, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.1951219512195124, | |
| "grad_norm": 2.224029018392997, | |
| "learning_rate": 9.531538935183249e-07, | |
| "loss": 2.5934, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.1951219512195124, | |
| "eval_loss": 2.5593512058258057, | |
| "eval_runtime": 18.9816, | |
| "eval_samples_per_second": 38.88, | |
| "eval_steps_per_second": 1.001, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "grad_norm": 2.2412373397938583, | |
| "learning_rate": 9.330127018922193e-07, | |
| "loss": 2.5543, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "eval_loss": 2.521718740463257, | |
| "eval_runtime": 20.215, | |
| "eval_samples_per_second": 36.508, | |
| "eval_steps_per_second": 0.94, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.682926829268293, | |
| "grad_norm": 2.209083408074763, | |
| "learning_rate": 9.095760221444959e-07, | |
| "loss": 2.513, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.682926829268293, | |
| "eval_loss": 2.4829368591308594, | |
| "eval_runtime": 19.4395, | |
| "eval_samples_per_second": 37.964, | |
| "eval_steps_per_second": 0.977, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.926829268292683, | |
| "grad_norm": 2.297068739774513, | |
| "learning_rate": 8.83022221559489e-07, | |
| "loss": 2.4712, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.926829268292683, | |
| "eval_loss": 2.446091890335083, | |
| "eval_runtime": 19.007, | |
| "eval_samples_per_second": 38.828, | |
| "eval_steps_per_second": 1.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.1707317073170733, | |
| "grad_norm": 2.3349163677591314, | |
| "learning_rate": 8.535533905932737e-07, | |
| "loss": 2.4365, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 3.1707317073170733, | |
| "eval_loss": 2.413790225982666, | |
| "eval_runtime": 20.2301, | |
| "eval_samples_per_second": 36.48, | |
| "eval_steps_per_second": 0.939, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 3.4146341463414633, | |
| "grad_norm": 2.285003975033001, | |
| "learning_rate": 8.213938048432696e-07, | |
| "loss": 2.4066, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.4146341463414633, | |
| "eval_loss": 2.3858845233917236, | |
| "eval_runtime": 18.3399, | |
| "eval_samples_per_second": 40.24, | |
| "eval_steps_per_second": 1.036, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.658536585365854, | |
| "grad_norm": 2.1725408410741203, | |
| "learning_rate": 7.86788218175523e-07, | |
| "loss": 2.375, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.658536585365854, | |
| "eval_loss": 2.3606066703796387, | |
| "eval_runtime": 19.9461, | |
| "eval_samples_per_second": 37.0, | |
| "eval_steps_per_second": 0.953, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.902439024390244, | |
| "grad_norm": 2.2674210698852133, | |
| "learning_rate": 7.5e-07, | |
| "loss": 2.3415, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.902439024390244, | |
| "eval_loss": 2.336864709854126, | |
| "eval_runtime": 18.4703, | |
| "eval_samples_per_second": 39.956, | |
| "eval_steps_per_second": 1.029, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.146341463414634, | |
| "grad_norm": 2.22603925687789, | |
| "learning_rate": 7.113091308703497e-07, | |
| "loss": 2.3225, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 4.146341463414634, | |
| "eval_loss": 2.3142693042755127, | |
| "eval_runtime": 19.8254, | |
| "eval_samples_per_second": 37.225, | |
| "eval_steps_per_second": 0.958, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 4.390243902439025, | |
| "grad_norm": 2.250461857867485, | |
| "learning_rate": 6.710100716628344e-07, | |
| "loss": 2.2989, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.390243902439025, | |
| "eval_loss": 2.292672634124756, | |
| "eval_runtime": 19.1763, | |
| "eval_samples_per_second": 38.485, | |
| "eval_steps_per_second": 0.991, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.634146341463414, | |
| "grad_norm": 2.2063059930566107, | |
| "learning_rate": 6.294095225512604e-07, | |
| "loss": 2.2748, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 4.634146341463414, | |
| "eval_loss": 2.2731637954711914, | |
| "eval_runtime": 18.4462, | |
| "eval_samples_per_second": 40.008, | |
| "eval_steps_per_second": 1.03, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 4.878048780487805, | |
| "grad_norm": 2.2114255470751205, | |
| "learning_rate": 5.868240888334652e-07, | |
| "loss": 2.2513, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.878048780487805, | |
| "eval_loss": 2.2562241554260254, | |
| "eval_runtime": 19.9755, | |
| "eval_samples_per_second": 36.945, | |
| "eval_steps_per_second": 0.951, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.121951219512195, | |
| "grad_norm": 2.1470949247310247, | |
| "learning_rate": 5.435778713738292e-07, | |
| "loss": 2.2401, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 5.121951219512195, | |
| "eval_loss": 2.241244077682495, | |
| "eval_runtime": 18.4347, | |
| "eval_samples_per_second": 40.033, | |
| "eval_steps_per_second": 1.031, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 5.365853658536586, | |
| "grad_norm": 2.2399669699712543, | |
| "learning_rate": 5e-07, | |
| "loss": 2.2172, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.365853658536586, | |
| "eval_loss": 2.228184461593628, | |
| "eval_runtime": 19.9167, | |
| "eval_samples_per_second": 37.054, | |
| "eval_steps_per_second": 0.954, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.609756097560975, | |
| "grad_norm": 2.248609346835798, | |
| "learning_rate": 4.5642212862617085e-07, | |
| "loss": 2.204, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.609756097560975, | |
| "eval_loss": 2.216791868209839, | |
| "eval_runtime": 18.6439, | |
| "eval_samples_per_second": 39.584, | |
| "eval_steps_per_second": 1.019, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.853658536585366, | |
| "grad_norm": 2.2843160705737393, | |
| "learning_rate": 4.131759111665348e-07, | |
| "loss": 2.1893, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.853658536585366, | |
| "eval_loss": 2.2068991661071777, | |
| "eval_runtime": 19.7257, | |
| "eval_samples_per_second": 37.413, | |
| "eval_steps_per_second": 0.963, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.097560975609756, | |
| "grad_norm": 2.198659146419105, | |
| "learning_rate": 3.7059047744873955e-07, | |
| "loss": 2.1784, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 6.097560975609756, | |
| "eval_loss": 2.1983890533447266, | |
| "eval_runtime": 20.2531, | |
| "eval_samples_per_second": 36.439, | |
| "eval_steps_per_second": 0.938, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 6.341463414634147, | |
| "grad_norm": 2.2746004739227996, | |
| "learning_rate": 3.2898992833716563e-07, | |
| "loss": 2.1646, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.341463414634147, | |
| "eval_loss": 2.191380500793457, | |
| "eval_runtime": 18.8387, | |
| "eval_samples_per_second": 39.175, | |
| "eval_steps_per_second": 1.009, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.585365853658536, | |
| "grad_norm": 2.182140726336947, | |
| "learning_rate": 2.8869086912965036e-07, | |
| "loss": 2.1673, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 6.585365853658536, | |
| "eval_loss": 2.185249090194702, | |
| "eval_runtime": 20.3689, | |
| "eval_samples_per_second": 36.232, | |
| "eval_steps_per_second": 0.933, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 6.829268292682927, | |
| "grad_norm": 2.2128884344288244, | |
| "learning_rate": 2.500000000000001e-07, | |
| "loss": 2.1555, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.829268292682927, | |
| "eval_loss": 2.180130958557129, | |
| "eval_runtime": 19.0502, | |
| "eval_samples_per_second": 38.74, | |
| "eval_steps_per_second": 0.997, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.073170731707317, | |
| "grad_norm": 2.102015386581095, | |
| "learning_rate": 2.1321178182447709e-07, | |
| "loss": 2.1599, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 7.073170731707317, | |
| "eval_loss": 2.175684690475464, | |
| "eval_runtime": 20.511, | |
| "eval_samples_per_second": 35.981, | |
| "eval_steps_per_second": 0.926, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 7.317073170731708, | |
| "grad_norm": 2.313350074598611, | |
| "learning_rate": 1.7860619515673032e-07, | |
| "loss": 2.145, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.317073170731708, | |
| "eval_loss": 2.1720824241638184, | |
| "eval_runtime": 18.8834, | |
| "eval_samples_per_second": 39.082, | |
| "eval_steps_per_second": 1.006, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.560975609756097, | |
| "grad_norm": 2.253715272870049, | |
| "learning_rate": 1.4644660940672627e-07, | |
| "loss": 2.1359, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 7.560975609756097, | |
| "eval_loss": 2.169234275817871, | |
| "eval_runtime": 19.3523, | |
| "eval_samples_per_second": 38.135, | |
| "eval_steps_per_second": 0.982, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 7.804878048780488, | |
| "grad_norm": 2.216308681434654, | |
| "learning_rate": 1.1697777844051104e-07, | |
| "loss": 2.1391, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.804878048780488, | |
| "eval_loss": 2.1668455600738525, | |
| "eval_runtime": 20.0953, | |
| "eval_samples_per_second": 36.725, | |
| "eval_steps_per_second": 0.945, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.048780487804878, | |
| "grad_norm": 2.2398335056684116, | |
| "learning_rate": 9.042397785550404e-08, | |
| "loss": 2.1274, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 8.048780487804878, | |
| "eval_loss": 2.1650218963623047, | |
| "eval_runtime": 18.6515, | |
| "eval_samples_per_second": 39.568, | |
| "eval_steps_per_second": 1.019, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 8.292682926829269, | |
| "grad_norm": 2.183896419200447, | |
| "learning_rate": 6.698729810778064e-08, | |
| "loss": 2.1342, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.292682926829269, | |
| "eval_loss": 2.163686752319336, | |
| "eval_runtime": 20.1658, | |
| "eval_samples_per_second": 36.597, | |
| "eval_steps_per_second": 0.942, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.536585365853659, | |
| "grad_norm": 2.260276175787966, | |
| "learning_rate": 4.684610648167503e-08, | |
| "loss": 2.1272, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 8.536585365853659, | |
| "eval_loss": 2.1627187728881836, | |
| "eval_runtime": 18.4069, | |
| "eval_samples_per_second": 40.094, | |
| "eval_steps_per_second": 1.032, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 8.78048780487805, | |
| "grad_norm": 2.2467910530017043, | |
| "learning_rate": 3.015368960704584e-08, | |
| "loss": 2.133, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.78048780487805, | |
| "eval_loss": 2.162067413330078, | |
| "eval_runtime": 19.691, | |
| "eval_samples_per_second": 37.479, | |
| "eval_steps_per_second": 0.965, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.024390243902438, | |
| "grad_norm": 2.2631951574285387, | |
| "learning_rate": 1.7037086855465898e-08, | |
| "loss": 2.1286, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 9.024390243902438, | |
| "eval_loss": 2.1616575717926025, | |
| "eval_runtime": 19.1797, | |
| "eval_samples_per_second": 38.478, | |
| "eval_steps_per_second": 0.991, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 9.268292682926829, | |
| "grad_norm": 2.2305961635805214, | |
| "learning_rate": 7.59612349389599e-09, | |
| "loss": 2.1296, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.268292682926829, | |
| "eval_loss": 2.161451578140259, | |
| "eval_runtime": 18.6125, | |
| "eval_samples_per_second": 39.651, | |
| "eval_steps_per_second": 1.021, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.512195121951219, | |
| "grad_norm": 2.209641020058134, | |
| "learning_rate": 1.9026509541272273e-09, | |
| "loss": 2.1256, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 9.512195121951219, | |
| "eval_loss": 2.161362409591675, | |
| "eval_runtime": 20.0337, | |
| "eval_samples_per_second": 36.838, | |
| "eval_steps_per_second": 0.948, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 9.75609756097561, | |
| "grad_norm": 2.2480648233150617, | |
| "learning_rate": 0.0, | |
| "loss": 2.1267, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.75609756097561, | |
| "eval_loss": 2.1613569259643555, | |
| "eval_runtime": 18.4232, | |
| "eval_samples_per_second": 40.058, | |
| "eval_steps_per_second": 1.031, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.75609756097561, | |
| "step": 200, | |
| "total_flos": 1.742036445167616e+16, | |
| "train_loss": 2.3644245743751524, | |
| "train_runtime": 7118.2054, | |
| "train_samples_per_second": 9.167, | |
| "train_steps_per_second": 0.028 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.742036445167616e+16, | |
| "train_batch_size": 10, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |