| { | |
| "best_metric": 4.30883264541626, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Mistral-7B-v0.3_metamath_default/checkpoint-572", | |
| "epoch": 0.9898744430943702, | |
| "eval_steps": 13, | |
| "global_step": 611, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0016200891049007696, | |
| "grad_norm": 31.97899627685547, | |
| "learning_rate": 2.3076923076923076e-05, | |
| "loss": 0.9728, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011340623734305387, | |
| "grad_norm": 25.20884132385254, | |
| "learning_rate": 0.00016153846153846153, | |
| "loss": 0.7439, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.021061158363710003, | |
| "eval_loss": 4.518104553222656, | |
| "eval_runtime": 13.0104, | |
| "eval_samples_per_second": 38.431, | |
| "eval_steps_per_second": 4.842, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.022681247468610773, | |
| "grad_norm": 9073.8154296875, | |
| "learning_rate": 0.0002999979709808197, | |
| "loss": 1.7362, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03402187120291616, | |
| "grad_norm": 35.26778793334961, | |
| "learning_rate": 0.0002998701612152596, | |
| "loss": 8.6977, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.042122316727420006, | |
| "eval_loss": 6.681583404541016, | |
| "eval_runtime": 84.9655, | |
| "eval_samples_per_second": 5.885, | |
| "eval_steps_per_second": 0.741, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.04536249493722155, | |
| "grad_norm": 12.110719680786133, | |
| "learning_rate": 0.0002995437011859465, | |
| "loss": 6.8454, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.056703118671526935, | |
| "grad_norm": 3.0557332038879395, | |
| "learning_rate": 0.00029901902360990936, | |
| "loss": 6.6139, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.06318347509113001, | |
| "eval_loss": 6.469002723693848, | |
| "eval_runtime": 82.1892, | |
| "eval_samples_per_second": 6.084, | |
| "eval_steps_per_second": 0.767, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.06804374240583232, | |
| "grad_norm": 3.914517879486084, | |
| "learning_rate": 0.00029829682393805085, | |
| "loss": 6.4334, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0793843661401377, | |
| "grad_norm": 15.051484107971191, | |
| "learning_rate": 0.0002973780594333385, | |
| "loss": 6.3559, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.08424463345484001, | |
| "eval_loss": 6.440323829650879, | |
| "eval_runtime": 12.6259, | |
| "eval_samples_per_second": 39.601, | |
| "eval_steps_per_second": 4.99, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.0907249898744431, | |
| "grad_norm": 4.897040843963623, | |
| "learning_rate": 0.00029626394790197025, | |
| "loss": 6.31, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.10206561360874848, | |
| "grad_norm": 5.669849872589111, | |
| "learning_rate": 0.00029495596607919305, | |
| "loss": 6.2844, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.10530579181855002, | |
| "eval_loss": 6.300673007965088, | |
| "eval_runtime": 83.0722, | |
| "eval_samples_per_second": 6.019, | |
| "eval_steps_per_second": 0.758, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.11340623734305387, | |
| "grad_norm": 21.01810073852539, | |
| "learning_rate": 0.00029345584767191685, | |
| "loss": 6.2212, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.12474686107735926, | |
| "grad_norm": 4.398944854736328, | |
| "learning_rate": 0.0002917655810607161, | |
| "loss": 6.1379, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.12636695018226002, | |
| "eval_loss": 6.0592122077941895, | |
| "eval_runtime": 91.4719, | |
| "eval_samples_per_second": 5.466, | |
| "eval_steps_per_second": 0.689, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.13608748481166463, | |
| "grad_norm": 3.180732250213623, | |
| "learning_rate": 0.0002898874066642667, | |
| "loss": 5.9832, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.14742810854597002, | |
| "grad_norm": 38.04954147338867, | |
| "learning_rate": 0.00028782381396971003, | |
| "loss": 5.9506, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.14742810854597002, | |
| "eval_loss": 5.914348602294922, | |
| "eval_runtime": 12.9694, | |
| "eval_samples_per_second": 38.552, | |
| "eval_steps_per_second": 4.858, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.1587687322802754, | |
| "grad_norm": 40.107906341552734, | |
| "learning_rate": 0.00028557753823288173, | |
| "loss": 5.8886, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.16848926690968002, | |
| "eval_loss": 5.989041805267334, | |
| "eval_runtime": 88.0067, | |
| "eval_samples_per_second": 5.681, | |
| "eval_steps_per_second": 0.716, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.1701093560145808, | |
| "grad_norm": 47.72571563720703, | |
| "learning_rate": 0.0002831515568527781, | |
| "loss": 5.8829, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.1814499797488862, | |
| "grad_norm": 8.360991477966309, | |
| "learning_rate": 0.00028054908542506627, | |
| "loss": 5.7387, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.18955042527339003, | |
| "eval_loss": 5.65593957901001, | |
| "eval_runtime": 82.1522, | |
| "eval_samples_per_second": 6.086, | |
| "eval_steps_per_second": 0.767, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.19279060348319157, | |
| "grad_norm": 21.80311393737793, | |
| "learning_rate": 0.00027777357347986823, | |
| "loss": 5.6862, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.20413122721749696, | |
| "grad_norm": 19.518041610717773, | |
| "learning_rate": 0.00027482869990946986, | |
| "loss": 5.672, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.21061158363710003, | |
| "eval_loss": 5.724035739898682, | |
| "eval_runtime": 12.9581, | |
| "eval_samples_per_second": 38.586, | |
| "eval_steps_per_second": 4.862, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.21547185095180235, | |
| "grad_norm": 20.255475997924805, | |
| "learning_rate": 0.0002717183680920135, | |
| "loss": 5.6345, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 0.22681247468610774, | |
| "grad_norm": 8.896872520446777, | |
| "learning_rate": 0.00026844670071763906, | |
| "loss": 5.5425, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.23167274200081003, | |
| "eval_loss": 5.5164642333984375, | |
| "eval_runtime": 90.7285, | |
| "eval_samples_per_second": 5.511, | |
| "eval_steps_per_second": 0.694, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.23815309842041313, | |
| "grad_norm": 22.878341674804688, | |
| "learning_rate": 0.00026501803432393037, | |
| "loss": 5.4354, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.24949372215471852, | |
| "grad_norm": 25.566158294677734, | |
| "learning_rate": 0.00026143691354791145, | |
| "loss": 5.473, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 0.25273390036452004, | |
| "eval_loss": 5.423782825469971, | |
| "eval_runtime": 86.2539, | |
| "eval_samples_per_second": 5.797, | |
| "eval_steps_per_second": 0.73, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.2608343458890239, | |
| "grad_norm": 11.549739837646484, | |
| "learning_rate": 0.00025770808510220956, | |
| "loss": 5.3333, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 0.27217496962332927, | |
| "grad_norm": 16.626855850219727, | |
| "learning_rate": 0.00025383649148337105, | |
| "loss": 5.328, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.27379505872823007, | |
| "eval_loss": 5.343407154083252, | |
| "eval_runtime": 13.0401, | |
| "eval_samples_per_second": 38.343, | |
| "eval_steps_per_second": 4.831, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 0.28351559335763465, | |
| "grad_norm": 34.26865005493164, | |
| "learning_rate": 0.0002498272644206695, | |
| "loss": 5.2502, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.29485621709194004, | |
| "grad_norm": 8.343174934387207, | |
| "learning_rate": 0.0002456857180740884, | |
| "loss": 5.241, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 0.29485621709194004, | |
| "eval_loss": 5.208798885345459, | |
| "eval_runtime": 77.0828, | |
| "eval_samples_per_second": 6.487, | |
| "eval_steps_per_second": 0.817, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 0.30619684082624543, | |
| "grad_norm": 29.43370819091797, | |
| "learning_rate": 0.0002414173419904956, | |
| "loss": 5.2491, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 0.3159173754556501, | |
| "eval_loss": 5.329117298126221, | |
| "eval_runtime": 87.9681, | |
| "eval_samples_per_second": 5.684, | |
| "eval_steps_per_second": 0.716, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.3175374645605508, | |
| "grad_norm": 41.14483642578125, | |
| "learning_rate": 0.00023702779382734566, | |
| "loss": 5.3369, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.3288780882948562, | |
| "grad_norm": 12.607691764831543, | |
| "learning_rate": 0.0002325228918535541, | |
| "loss": 5.2994, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 0.33697853381936005, | |
| "eval_loss": 5.168725490570068, | |
| "eval_runtime": 13.0279, | |
| "eval_samples_per_second": 38.379, | |
| "eval_steps_per_second": 4.836, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.3402187120291616, | |
| "grad_norm": 13.904136657714844, | |
| "learning_rate": 0.00022790860723748442, | |
| "loss": 5.1467, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.351559335763467, | |
| "grad_norm": 27.545360565185547, | |
| "learning_rate": 0.00022319105613226921, | |
| "loss": 5.1595, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 0.3580396921830701, | |
| "eval_loss": 5.07974910736084, | |
| "eval_runtime": 93.1844, | |
| "eval_samples_per_second": 5.366, | |
| "eval_steps_per_second": 0.676, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 0.3628999594977724, | |
| "grad_norm": 9.475492477416992, | |
| "learning_rate": 0.00021837649156895706, | |
| "loss": 5.1084, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.37424058323207776, | |
| "grad_norm": 9.638204574584961, | |
| "learning_rate": 0.00021347129516822945, | |
| "loss": 5.0592, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 0.37910085054678005, | |
| "eval_loss": 5.000487327575684, | |
| "eval_runtime": 87.2972, | |
| "eval_samples_per_second": 5.728, | |
| "eval_steps_per_second": 0.722, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 0.38558120696638315, | |
| "grad_norm": 19.830158233642578, | |
| "learning_rate": 0.00020848196868167505, | |
| "loss": 5.0021, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 0.39692183070068854, | |
| "grad_norm": 40.01640319824219, | |
| "learning_rate": 0.000203415125373832, | |
| "loss": 4.9674, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.4001620089104901, | |
| "eval_loss": 4.952455520629883, | |
| "eval_runtime": 13.0727, | |
| "eval_samples_per_second": 38.248, | |
| "eval_steps_per_second": 4.819, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 0.4082624544349939, | |
| "grad_norm": 45.92347717285156, | |
| "learning_rate": 0.00019827748125642242, | |
| "loss": 4.9861, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.4196030781692993, | |
| "grad_norm": 7.856499195098877, | |
| "learning_rate": 0.0001930758461863965, | |
| "loss": 4.9663, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 0.42122316727420006, | |
| "eval_loss": 4.970428943634033, | |
| "eval_runtime": 79.1825, | |
| "eval_samples_per_second": 6.315, | |
| "eval_steps_per_second": 0.796, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.4309437019036047, | |
| "grad_norm": 45.39262771606445, | |
| "learning_rate": 0.0001878171148395872, | |
| "loss": 4.9635, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 0.4422843256379101, | |
| "grad_norm": 49.14884567260742, | |
| "learning_rate": 0.00018250825757193848, | |
| "loss": 5.0169, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 0.4422843256379101, | |
| "eval_loss": 4.971817970275879, | |
| "eval_runtime": 80.0188, | |
| "eval_samples_per_second": 6.249, | |
| "eval_steps_per_second": 0.787, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 0.4536249493722155, | |
| "grad_norm": 13.595600128173828, | |
| "learning_rate": 0.0001771563111804211, | |
| "loss": 4.9333, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.46334548400162007, | |
| "eval_loss": 4.827688694000244, | |
| "eval_runtime": 13.1128, | |
| "eval_samples_per_second": 38.131, | |
| "eval_steps_per_second": 4.804, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 0.46496557310652087, | |
| "grad_norm": 21.398765563964844, | |
| "learning_rate": 0.0001717683695758819, | |
| "loss": 4.7715, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 0.47630619684082626, | |
| "grad_norm": 17.379562377929688, | |
| "learning_rate": 0.00016635157438018983, | |
| "loss": 4.8687, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 0.4844066423653301, | |
| "eval_loss": 4.813088893890381, | |
| "eval_runtime": 90.9286, | |
| "eval_samples_per_second": 5.499, | |
| "eval_steps_per_second": 0.693, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 0.48764682057513165, | |
| "grad_norm": 36.88113021850586, | |
| "learning_rate": 0.0001609131054601416, | |
| "loss": 4.8246, | |
| "step": 301 | |
| }, | |
| { | |
| "epoch": 0.49898744430943703, | |
| "grad_norm": 27.84221076965332, | |
| "learning_rate": 0.00015546017141067432, | |
| "loss": 4.7215, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.5054678007290401, | |
| "eval_loss": 4.7605767250061035, | |
| "eval_runtime": 93.5587, | |
| "eval_samples_per_second": 5.344, | |
| "eval_steps_per_second": 0.673, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.5103280680437424, | |
| "grad_norm": 31.899227142333984, | |
| "learning_rate": 0.00015, | |
| "loss": 4.7799, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.5216686917780478, | |
| "grad_norm": 33.652374267578125, | |
| "learning_rate": 0.0001445398285893257, | |
| "loss": 4.7602, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 0.5265289590927501, | |
| "eval_loss": 4.704898834228516, | |
| "eval_runtime": 13.0792, | |
| "eval_samples_per_second": 38.229, | |
| "eval_steps_per_second": 4.817, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.5330093155123532, | |
| "grad_norm": 54.98084259033203, | |
| "learning_rate": 0.0001390868945398584, | |
| "loss": 4.6746, | |
| "step": 329 | |
| }, | |
| { | |
| "epoch": 0.5443499392466585, | |
| "grad_norm": 29.5, | |
| "learning_rate": 0.00013364842561981014, | |
| "loss": 4.7033, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.5475901174564601, | |
| "eval_loss": 4.78165340423584, | |
| "eval_runtime": 92.9113, | |
| "eval_samples_per_second": 5.381, | |
| "eval_steps_per_second": 0.678, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 0.555690562980964, | |
| "grad_norm": 16.196989059448242, | |
| "learning_rate": 0.00012823163042411807, | |
| "loss": 4.6363, | |
| "step": 343 | |
| }, | |
| { | |
| "epoch": 0.5670311867152693, | |
| "grad_norm": 22.850570678710938, | |
| "learning_rate": 0.0001228436888195789, | |
| "loss": 4.7179, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.56865127582017, | |
| "eval_loss": 4.642824172973633, | |
| "eval_runtime": 79.0871, | |
| "eval_samples_per_second": 6.322, | |
| "eval_steps_per_second": 0.797, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 0.5783718104495748, | |
| "grad_norm": 19.261425018310547, | |
| "learning_rate": 0.00011749174242806152, | |
| "loss": 4.6463, | |
| "step": 357 | |
| }, | |
| { | |
| "epoch": 0.5897124341838801, | |
| "grad_norm": 21.468109130859375, | |
| "learning_rate": 0.00011218288516041279, | |
| "loss": 4.6525, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.5897124341838801, | |
| "eval_loss": 4.596373081207275, | |
| "eval_runtime": 13.1407, | |
| "eval_samples_per_second": 38.05, | |
| "eval_steps_per_second": 4.794, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.6010530579181855, | |
| "grad_norm": 21.91275405883789, | |
| "learning_rate": 0.00010692415381360349, | |
| "loss": 4.5923, | |
| "step": 371 | |
| }, | |
| { | |
| "epoch": 0.6107735925475901, | |
| "eval_loss": 4.560819149017334, | |
| "eval_runtime": 78.8178, | |
| "eval_samples_per_second": 6.344, | |
| "eval_steps_per_second": 0.799, | |
| "step": 377 | |
| }, | |
| { | |
| "epoch": 0.6123936816524909, | |
| "grad_norm": 33.42408752441406, | |
| "learning_rate": 0.00010172251874357757, | |
| "loss": 4.6014, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 0.6237343053867963, | |
| "grad_norm": 24.099485397338867, | |
| "learning_rate": 9.658487462616794e-05, | |
| "loss": 4.5936, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.6318347509113001, | |
| "eval_loss": 4.567561626434326, | |
| "eval_runtime": 88.6357, | |
| "eval_samples_per_second": 5.641, | |
| "eval_steps_per_second": 0.711, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.6350749291211016, | |
| "grad_norm": 19.200098037719727, | |
| "learning_rate": 9.151803131832493e-05, | |
| "loss": 4.5747, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 0.6464155528554071, | |
| "grad_norm": 32.721920013427734, | |
| "learning_rate": 8.652870483177049e-05, | |
| "loss": 4.5142, | |
| "step": 399 | |
| }, | |
| { | |
| "epoch": 0.6528959092750102, | |
| "eval_loss": 4.501616954803467, | |
| "eval_runtime": 13.1244, | |
| "eval_samples_per_second": 38.097, | |
| "eval_steps_per_second": 4.8, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 0.6577561765897124, | |
| "grad_norm": 18.631881713867188, | |
| "learning_rate": 8.162350843104291e-05, | |
| "loss": 4.4781, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 0.6690968003240179, | |
| "grad_norm": 9.993534088134766, | |
| "learning_rate": 7.680894386773072e-05, | |
| "loss": 4.4717, | |
| "step": 413 | |
| }, | |
| { | |
| "epoch": 0.6739570676387201, | |
| "eval_loss": 4.4421563148498535, | |
| "eval_runtime": 92.2744, | |
| "eval_samples_per_second": 5.419, | |
| "eval_steps_per_second": 0.683, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 0.6804374240583232, | |
| "grad_norm": 54.20478820800781, | |
| "learning_rate": 7.209139276251558e-05, | |
| "loss": 4.4965, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.6917780477926286, | |
| "grad_norm": 8.450617790222168, | |
| "learning_rate": 6.747710814644589e-05, | |
| "loss": 4.5539, | |
| "step": 427 | |
| }, | |
| { | |
| "epoch": 0.6950182260024301, | |
| "eval_loss": 4.517686367034912, | |
| "eval_runtime": 92.691, | |
| "eval_samples_per_second": 5.394, | |
| "eval_steps_per_second": 0.68, | |
| "step": 429 | |
| }, | |
| { | |
| "epoch": 0.703118671526934, | |
| "grad_norm": 15.6026029586792, | |
| "learning_rate": 6.297220617265435e-05, | |
| "loss": 4.4799, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 0.7144592952612394, | |
| "grad_norm": 43.13425064086914, | |
| "learning_rate": 5.858265800950438e-05, | |
| "loss": 4.5129, | |
| "step": 441 | |
| }, | |
| { | |
| "epoch": 0.7160793843661402, | |
| "eval_loss": 4.539721488952637, | |
| "eval_runtime": 13.1249, | |
| "eval_samples_per_second": 38.095, | |
| "eval_steps_per_second": 4.8, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 0.7257999189955447, | |
| "grad_norm": 23.415088653564453, | |
| "learning_rate": 5.4314281925911634e-05, | |
| "loss": 4.4859, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 0.7371405427298502, | |
| "grad_norm": 25.57401466369629, | |
| "learning_rate": 5.0172735579330526e-05, | |
| "loss": 4.4162, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.7371405427298502, | |
| "eval_loss": 4.4049577713012695, | |
| "eval_runtime": 83.3117, | |
| "eval_samples_per_second": 6.002, | |
| "eval_steps_per_second": 0.756, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.7484811664641555, | |
| "grad_norm": 18.13247299194336, | |
| "learning_rate": 4.616350851662895e-05, | |
| "loss": 4.4328, | |
| "step": 462 | |
| }, | |
| { | |
| "epoch": 0.7582017010935601, | |
| "eval_loss": 4.424379348754883, | |
| "eval_runtime": 78.9244, | |
| "eval_samples_per_second": 6.335, | |
| "eval_steps_per_second": 0.798, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 0.759821790198461, | |
| "grad_norm": 31.204004287719727, | |
| "learning_rate": 4.229191489779047e-05, | |
| "loss": 4.3971, | |
| "step": 469 | |
| }, | |
| { | |
| "epoch": 0.7711624139327663, | |
| "grad_norm": 34.741661071777344, | |
| "learning_rate": 3.8563086452088506e-05, | |
| "loss": 4.3949, | |
| "step": 476 | |
| }, | |
| { | |
| "epoch": 0.7792628594572701, | |
| "eval_loss": 4.401327133178711, | |
| "eval_runtime": 13.1531, | |
| "eval_samples_per_second": 38.014, | |
| "eval_steps_per_second": 4.79, | |
| "step": 481 | |
| }, | |
| { | |
| "epoch": 0.7825030376670717, | |
| "grad_norm": 30.405717849731445, | |
| "learning_rate": 3.498196567606959e-05, | |
| "loss": 4.3752, | |
| "step": 483 | |
| }, | |
| { | |
| "epoch": 0.7938436614013771, | |
| "grad_norm": 27.06761932373047, | |
| "learning_rate": 3.1553299282360966e-05, | |
| "loss": 4.3946, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.8003240178209802, | |
| "eval_loss": 4.372068405151367, | |
| "eval_runtime": 95.0407, | |
| "eval_samples_per_second": 5.261, | |
| "eval_steps_per_second": 0.663, | |
| "step": 494 | |
| }, | |
| { | |
| "epoch": 0.8051842851356824, | |
| "grad_norm": 11.737322807312012, | |
| "learning_rate": 2.828163190798644e-05, | |
| "loss": 4.3378, | |
| "step": 497 | |
| }, | |
| { | |
| "epoch": 0.8165249088699879, | |
| "grad_norm": 27.00642204284668, | |
| "learning_rate": 2.5171300090530106e-05, | |
| "loss": 4.393, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 0.8213851761846902, | |
| "eval_loss": 4.358621120452881, | |
| "eval_runtime": 92.1383, | |
| "eval_samples_per_second": 5.427, | |
| "eval_steps_per_second": 0.684, | |
| "step": 507 | |
| }, | |
| { | |
| "epoch": 0.8278655326042932, | |
| "grad_norm": 22.39855194091797, | |
| "learning_rate": 2.2226426520131734e-05, | |
| "loss": 4.3436, | |
| "step": 511 | |
| }, | |
| { | |
| "epoch": 0.8392061563385986, | |
| "grad_norm": 15.117547988891602, | |
| "learning_rate": 1.9450914574933725e-05, | |
| "loss": 4.3872, | |
| "step": 518 | |
| }, | |
| { | |
| "epoch": 0.8424463345484001, | |
| "eval_loss": 4.355201244354248, | |
| "eval_runtime": 13.114, | |
| "eval_samples_per_second": 38.127, | |
| "eval_steps_per_second": 4.804, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.850546780072904, | |
| "grad_norm": 22.19342041015625, | |
| "learning_rate": 1.6848443147221828e-05, | |
| "loss": 4.3434, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.8618874038072094, | |
| "grad_norm": 21.237140655517578, | |
| "learning_rate": 1.4422461767118233e-05, | |
| "loss": 4.3787, | |
| "step": 532 | |
| }, | |
| { | |
| "epoch": 0.8635074929121102, | |
| "eval_loss": 4.404301166534424, | |
| "eval_runtime": 89.2424, | |
| "eval_samples_per_second": 5.603, | |
| "eval_steps_per_second": 0.706, | |
| "step": 533 | |
| }, | |
| { | |
| "epoch": 0.8732280275415147, | |
| "grad_norm": 11.590981483459473, | |
| "learning_rate": 1.2176186030289936e-05, | |
| "loss": 4.3491, | |
| "step": 539 | |
| }, | |
| { | |
| "epoch": 0.8845686512758202, | |
| "grad_norm": 7.868985652923584, | |
| "learning_rate": 1.011259333573326e-05, | |
| "loss": 4.3477, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 0.8845686512758202, | |
| "eval_loss": 4.3536529541015625, | |
| "eval_runtime": 77.8001, | |
| "eval_samples_per_second": 6.427, | |
| "eval_steps_per_second": 0.81, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 0.8959092750101255, | |
| "grad_norm": 22.794200897216797, | |
| "learning_rate": 8.234418939283866e-06, | |
| "loss": 4.3957, | |
| "step": 553 | |
| }, | |
| { | |
| "epoch": 0.9056298096395302, | |
| "eval_loss": 4.333388805389404, | |
| "eval_runtime": 13.1267, | |
| "eval_samples_per_second": 38.09, | |
| "eval_steps_per_second": 4.799, | |
| "step": 559 | |
| }, | |
| { | |
| "epoch": 0.907249898744431, | |
| "grad_norm": 11.550942420959473, | |
| "learning_rate": 6.544152328083152e-06, | |
| "loss": 4.3827, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.9185905224787363, | |
| "grad_norm": 11.666648864746094, | |
| "learning_rate": 5.044033920806933e-06, | |
| "loss": 4.3634, | |
| "step": 567 | |
| }, | |
| { | |
| "epoch": 0.9266909680032401, | |
| "eval_loss": 4.30883264541626, | |
| "eval_runtime": 89.3126, | |
| "eval_samples_per_second": 5.598, | |
| "eval_steps_per_second": 0.705, | |
| "step": 572 | |
| }, | |
| { | |
| "epoch": 0.9299311462130417, | |
| "grad_norm": 13.062474250793457, | |
| "learning_rate": 3.7360520980297514e-06, | |
| "loss": 4.3485, | |
| "step": 574 | |
| }, | |
| { | |
| "epoch": 0.9412717699473471, | |
| "grad_norm": 10.294334411621094, | |
| "learning_rate": 2.6219405666614402e-06, | |
| "loss": 4.2898, | |
| "step": 581 | |
| }, | |
| { | |
| "epoch": 0.9477521263669502, | |
| "eval_loss": 4.334242820739746, | |
| "eval_runtime": 92.3717, | |
| "eval_samples_per_second": 5.413, | |
| "eval_steps_per_second": 0.682, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 0.9526123936816525, | |
| "grad_norm": 8.134040832519531, | |
| "learning_rate": 1.7031760619491353e-06, | |
| "loss": 4.3431, | |
| "step": 588 | |
| }, | |
| { | |
| "epoch": 0.9639530174159578, | |
| "grad_norm": 9.691034317016602, | |
| "learning_rate": 9.809763900905875e-07, | |
| "loss": 4.355, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 0.9688132847306602, | |
| "eval_loss": 4.32021951675415, | |
| "eval_runtime": 13.1964, | |
| "eval_samples_per_second": 37.889, | |
| "eval_steps_per_second": 4.774, | |
| "step": 598 | |
| }, | |
| { | |
| "epoch": 0.9752936411502633, | |
| "grad_norm": 14.163640975952148, | |
| "learning_rate": 4.562988140535073e-07, | |
| "loss": 4.327, | |
| "step": 602 | |
| }, | |
| { | |
| "epoch": 0.9866342648845686, | |
| "grad_norm": 8.559078216552734, | |
| "learning_rate": 1.298387847403437e-07, | |
| "loss": 4.3331, | |
| "step": 609 | |
| }, | |
| { | |
| "epoch": 0.9898744430943702, | |
| "eval_loss": 4.323936462402344, | |
| "eval_runtime": 86.9395, | |
| "eval_samples_per_second": 5.751, | |
| "eval_steps_per_second": 0.725, | |
| "step": 611 | |
| } | |
| ], | |
| "logging_steps": 7, | |
| "max_steps": 617, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.92558921542271e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |