|
{ |
|
"best_metric": 0.4760262072086334, |
|
"best_model_checkpoint": "/home/datta0/models/lora_final/Meta-Llama-3-8B_metamath_ortho/checkpoint-611", |
|
"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": 13.657631874084473, |
|
"learning_rate": 2.3076923076923076e-05, |
|
"loss": 1.2159, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.011340623734305387, |
|
"grad_norm": 4.022308349609375, |
|
"learning_rate": 0.00016153846153846153, |
|
"loss": 0.9348, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 0.021061158363710003, |
|
"eval_loss": 0.7025884389877319, |
|
"eval_runtime": 12.1198, |
|
"eval_samples_per_second": 41.255, |
|
"eval_steps_per_second": 5.198, |
|
"step": 13 |
|
}, |
|
{ |
|
"epoch": 0.022681247468610773, |
|
"grad_norm": 2.0974254608154297, |
|
"learning_rate": 0.0002999979709808197, |
|
"loss": 0.6798, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 0.03402187120291616, |
|
"grad_norm": 2.313894033432007, |
|
"learning_rate": 0.0002998701612152596, |
|
"loss": 0.6609, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 0.042122316727420006, |
|
"eval_loss": 0.6959722638130188, |
|
"eval_runtime": 76.7856, |
|
"eval_samples_per_second": 6.512, |
|
"eval_steps_per_second": 0.82, |
|
"step": 26 |
|
}, |
|
{ |
|
"epoch": 0.04536249493722155, |
|
"grad_norm": 2.249401569366455, |
|
"learning_rate": 0.0002995437011859465, |
|
"loss": 0.6709, |
|
"step": 28 |
|
}, |
|
{ |
|
"epoch": 0.056703118671526935, |
|
"grad_norm": 1.9721312522888184, |
|
"learning_rate": 0.00029901902360990936, |
|
"loss": 0.6695, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 0.06318347509113001, |
|
"eval_loss": 0.6784710884094238, |
|
"eval_runtime": 79.9084, |
|
"eval_samples_per_second": 6.257, |
|
"eval_steps_per_second": 0.788, |
|
"step": 39 |
|
}, |
|
{ |
|
"epoch": 0.06804374240583232, |
|
"grad_norm": 2.082033395767212, |
|
"learning_rate": 0.00029829682393805085, |
|
"loss": 0.6472, |
|
"step": 42 |
|
}, |
|
{ |
|
"epoch": 0.0793843661401377, |
|
"grad_norm": 2.075465202331543, |
|
"learning_rate": 0.0002973780594333385, |
|
"loss": 0.6578, |
|
"step": 49 |
|
}, |
|
{ |
|
"epoch": 0.08424463345484001, |
|
"eval_loss": 0.6770208477973938, |
|
"eval_runtime": 12.0924, |
|
"eval_samples_per_second": 41.348, |
|
"eval_steps_per_second": 5.21, |
|
"step": 52 |
|
}, |
|
{ |
|
"epoch": 0.0907249898744431, |
|
"grad_norm": 2.3162450790405273, |
|
"learning_rate": 0.00029626394790197025, |
|
"loss": 0.651, |
|
"step": 56 |
|
}, |
|
{ |
|
"epoch": 0.10206561360874848, |
|
"grad_norm": 2.016871929168701, |
|
"learning_rate": 0.00029495596607919305, |
|
"loss": 0.6222, |
|
"step": 63 |
|
}, |
|
{ |
|
"epoch": 0.10530579181855002, |
|
"eval_loss": 0.6747815012931824, |
|
"eval_runtime": 57.4761, |
|
"eval_samples_per_second": 8.699, |
|
"eval_steps_per_second": 1.096, |
|
"step": 65 |
|
}, |
|
{ |
|
"epoch": 0.11340623734305387, |
|
"grad_norm": 2.5055863857269287, |
|
"learning_rate": 0.00029345584767191685, |
|
"loss": 0.6403, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.12474686107735926, |
|
"grad_norm": 2.1047961711883545, |
|
"learning_rate": 0.0002917655810607161, |
|
"loss": 0.6331, |
|
"step": 77 |
|
}, |
|
{ |
|
"epoch": 0.12636695018226002, |
|
"eval_loss": 0.6662291288375854, |
|
"eval_runtime": 71.288, |
|
"eval_samples_per_second": 7.014, |
|
"eval_steps_per_second": 0.884, |
|
"step": 78 |
|
}, |
|
{ |
|
"epoch": 0.13608748481166463, |
|
"grad_norm": 1.7916154861450195, |
|
"learning_rate": 0.0002898874066642667, |
|
"loss": 0.6259, |
|
"step": 84 |
|
}, |
|
{ |
|
"epoch": 0.14742810854597002, |
|
"grad_norm": 2.046363592147827, |
|
"learning_rate": 0.00028782381396971003, |
|
"loss": 0.6413, |
|
"step": 91 |
|
}, |
|
{ |
|
"epoch": 0.14742810854597002, |
|
"eval_loss": 0.6609434485435486, |
|
"eval_runtime": 12.131, |
|
"eval_samples_per_second": 41.217, |
|
"eval_steps_per_second": 5.193, |
|
"step": 91 |
|
}, |
|
{ |
|
"epoch": 0.1587687322802754, |
|
"grad_norm": 1.9861701726913452, |
|
"learning_rate": 0.00028557753823288173, |
|
"loss": 0.631, |
|
"step": 98 |
|
}, |
|
{ |
|
"epoch": 0.16848926690968002, |
|
"eval_loss": 0.6538010239601135, |
|
"eval_runtime": 80.9566, |
|
"eval_samples_per_second": 6.176, |
|
"eval_steps_per_second": 0.778, |
|
"step": 104 |
|
}, |
|
{ |
|
"epoch": 0.1701093560145808, |
|
"grad_norm": 1.846071720123291, |
|
"learning_rate": 0.0002831515568527781, |
|
"loss": 0.6475, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 0.1814499797488862, |
|
"grad_norm": 2.3317582607269287, |
|
"learning_rate": 0.00028054908542506627, |
|
"loss": 0.6115, |
|
"step": 112 |
|
}, |
|
{ |
|
"epoch": 0.18955042527339003, |
|
"eval_loss": 0.6490179896354675, |
|
"eval_runtime": 82.3645, |
|
"eval_samples_per_second": 6.071, |
|
"eval_steps_per_second": 0.765, |
|
"step": 117 |
|
}, |
|
{ |
|
"epoch": 0.19279060348319157, |
|
"grad_norm": 1.7408918142318726, |
|
"learning_rate": 0.00027777357347986823, |
|
"loss": 0.6253, |
|
"step": 119 |
|
}, |
|
{ |
|
"epoch": 0.20413122721749696, |
|
"grad_norm": 1.6225473880767822, |
|
"learning_rate": 0.00027482869990946986, |
|
"loss": 0.6097, |
|
"step": 126 |
|
}, |
|
{ |
|
"epoch": 0.21061158363710003, |
|
"eval_loss": 0.6458165049552917, |
|
"eval_runtime": 12.1355, |
|
"eval_samples_per_second": 41.201, |
|
"eval_steps_per_second": 5.191, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.21547185095180235, |
|
"grad_norm": 2.08821964263916, |
|
"learning_rate": 0.0002717183680920135, |
|
"loss": 0.6016, |
|
"step": 133 |
|
}, |
|
{ |
|
"epoch": 0.22681247468610774, |
|
"grad_norm": 2.0556275844573975, |
|
"learning_rate": 0.00026844670071763906, |
|
"loss": 0.6052, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.23167274200081003, |
|
"eval_loss": 0.6532984375953674, |
|
"eval_runtime": 80.8235, |
|
"eval_samples_per_second": 6.186, |
|
"eval_steps_per_second": 0.779, |
|
"step": 143 |
|
}, |
|
{ |
|
"epoch": 0.23815309842041313, |
|
"grad_norm": 2.4191389083862305, |
|
"learning_rate": 0.00026501803432393037, |
|
"loss": 0.6318, |
|
"step": 147 |
|
}, |
|
{ |
|
"epoch": 0.24949372215471852, |
|
"grad_norm": 2.222801923751831, |
|
"learning_rate": 0.00026143691354791145, |
|
"loss": 0.6131, |
|
"step": 154 |
|
}, |
|
{ |
|
"epoch": 0.25273390036452004, |
|
"eval_loss": 0.6334775686264038, |
|
"eval_runtime": 82.9349, |
|
"eval_samples_per_second": 6.029, |
|
"eval_steps_per_second": 0.76, |
|
"step": 156 |
|
}, |
|
{ |
|
"epoch": 0.2608343458890239, |
|
"grad_norm": 1.741701364517212, |
|
"learning_rate": 0.00025770808510220956, |
|
"loss": 0.6015, |
|
"step": 161 |
|
}, |
|
{ |
|
"epoch": 0.27217496962332927, |
|
"grad_norm": 1.7539088726043701, |
|
"learning_rate": 0.00025383649148337105, |
|
"loss": 0.6012, |
|
"step": 168 |
|
}, |
|
{ |
|
"epoch": 0.27379505872823007, |
|
"eval_loss": 0.6165497303009033, |
|
"eval_runtime": 12.1274, |
|
"eval_samples_per_second": 41.229, |
|
"eval_steps_per_second": 5.195, |
|
"step": 169 |
|
}, |
|
{ |
|
"epoch": 0.28351559335763465, |
|
"grad_norm": 2.0264108180999756, |
|
"learning_rate": 0.0002498272644206695, |
|
"loss": 0.5974, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 0.29485621709194004, |
|
"grad_norm": 1.7606844902038574, |
|
"learning_rate": 0.0002456857180740884, |
|
"loss": 0.6002, |
|
"step": 182 |
|
}, |
|
{ |
|
"epoch": 0.29485621709194004, |
|
"eval_loss": 0.620435893535614, |
|
"eval_runtime": 85.68, |
|
"eval_samples_per_second": 5.836, |
|
"eval_steps_per_second": 0.735, |
|
"step": 182 |
|
}, |
|
{ |
|
"epoch": 0.30619684082624543, |
|
"grad_norm": 1.691292643547058, |
|
"learning_rate": 0.0002414173419904956, |
|
"loss": 0.5849, |
|
"step": 189 |
|
}, |
|
{ |
|
"epoch": 0.3159173754556501, |
|
"eval_loss": 0.6220102906227112, |
|
"eval_runtime": 82.9019, |
|
"eval_samples_per_second": 6.031, |
|
"eval_steps_per_second": 0.76, |
|
"step": 195 |
|
}, |
|
{ |
|
"epoch": 0.3175374645605508, |
|
"grad_norm": 1.944412350654602, |
|
"learning_rate": 0.00023702779382734566, |
|
"loss": 0.5901, |
|
"step": 196 |
|
}, |
|
{ |
|
"epoch": 0.3288780882948562, |
|
"grad_norm": 1.9867457151412964, |
|
"learning_rate": 0.0002325228918535541, |
|
"loss": 0.5689, |
|
"step": 203 |
|
}, |
|
{ |
|
"epoch": 0.33697853381936005, |
|
"eval_loss": 0.6159170866012573, |
|
"eval_runtime": 12.1326, |
|
"eval_samples_per_second": 41.211, |
|
"eval_steps_per_second": 5.193, |
|
"step": 208 |
|
}, |
|
{ |
|
"epoch": 0.3402187120291616, |
|
"grad_norm": 1.955578327178955, |
|
"learning_rate": 0.00022790860723748442, |
|
"loss": 0.5924, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.351559335763467, |
|
"grad_norm": 1.8735487461090088, |
|
"learning_rate": 0.00022319105613226921, |
|
"loss": 0.5781, |
|
"step": 217 |
|
}, |
|
{ |
|
"epoch": 0.3580396921830701, |
|
"eval_loss": 0.6110228300094604, |
|
"eval_runtime": 81.2619, |
|
"eval_samples_per_second": 6.153, |
|
"eval_steps_per_second": 0.775, |
|
"step": 221 |
|
}, |
|
{ |
|
"epoch": 0.3628999594977724, |
|
"grad_norm": 2.644158363342285, |
|
"learning_rate": 0.00021837649156895706, |
|
"loss": 0.5891, |
|
"step": 224 |
|
}, |
|
{ |
|
"epoch": 0.37424058323207776, |
|
"grad_norm": 1.511711597442627, |
|
"learning_rate": 0.00021347129516822945, |
|
"loss": 0.5765, |
|
"step": 231 |
|
}, |
|
{ |
|
"epoch": 0.37910085054678005, |
|
"eval_loss": 0.6027255654335022, |
|
"eval_runtime": 82.4543, |
|
"eval_samples_per_second": 6.064, |
|
"eval_steps_per_second": 0.764, |
|
"step": 234 |
|
}, |
|
{ |
|
"epoch": 0.38558120696638315, |
|
"grad_norm": 1.8768197298049927, |
|
"learning_rate": 0.00020848196868167505, |
|
"loss": 0.5932, |
|
"step": 238 |
|
}, |
|
{ |
|
"epoch": 0.39692183070068854, |
|
"grad_norm": 1.6268372535705566, |
|
"learning_rate": 0.000203415125373832, |
|
"loss": 0.5899, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 0.4001620089104901, |
|
"eval_loss": 0.5983073711395264, |
|
"eval_runtime": 12.1671, |
|
"eval_samples_per_second": 41.094, |
|
"eval_steps_per_second": 5.178, |
|
"step": 247 |
|
}, |
|
{ |
|
"epoch": 0.4082624544349939, |
|
"grad_norm": 2.1390023231506348, |
|
"learning_rate": 0.00019827748125642242, |
|
"loss": 0.5816, |
|
"step": 252 |
|
}, |
|
{ |
|
"epoch": 0.4196030781692993, |
|
"grad_norm": 1.6386767625808716, |
|
"learning_rate": 0.0001930758461863965, |
|
"loss": 0.5638, |
|
"step": 259 |
|
}, |
|
{ |
|
"epoch": 0.42122316727420006, |
|
"eval_loss": 0.5904666781425476, |
|
"eval_runtime": 81.2985, |
|
"eval_samples_per_second": 6.15, |
|
"eval_steps_per_second": 0.775, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.4309437019036047, |
|
"grad_norm": 1.5721111297607422, |
|
"learning_rate": 0.0001878171148395872, |
|
"loss": 0.5694, |
|
"step": 266 |
|
}, |
|
{ |
|
"epoch": 0.4422843256379101, |
|
"grad_norm": 1.6830506324768066, |
|
"learning_rate": 0.00018250825757193848, |
|
"loss": 0.5716, |
|
"step": 273 |
|
}, |
|
{ |
|
"epoch": 0.4422843256379101, |
|
"eval_loss": 0.5874103903770447, |
|
"eval_runtime": 81.2386, |
|
"eval_samples_per_second": 6.155, |
|
"eval_steps_per_second": 0.775, |
|
"step": 273 |
|
}, |
|
{ |
|
"epoch": 0.4536249493722155, |
|
"grad_norm": 1.5602617263793945, |
|
"learning_rate": 0.0001771563111804211, |
|
"loss": 0.5729, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.46334548400162007, |
|
"eval_loss": 0.5809030532836914, |
|
"eval_runtime": 12.1549, |
|
"eval_samples_per_second": 41.136, |
|
"eval_steps_per_second": 5.183, |
|
"step": 286 |
|
}, |
|
{ |
|
"epoch": 0.46496557310652087, |
|
"grad_norm": 1.4982717037200928, |
|
"learning_rate": 0.0001717683695758819, |
|
"loss": 0.5542, |
|
"step": 287 |
|
}, |
|
{ |
|
"epoch": 0.47630619684082626, |
|
"grad_norm": 1.5617115497589111, |
|
"learning_rate": 0.00016635157438018983, |
|
"loss": 0.5691, |
|
"step": 294 |
|
}, |
|
{ |
|
"epoch": 0.4844066423653301, |
|
"eval_loss": 0.5728867650032043, |
|
"eval_runtime": 89.8515, |
|
"eval_samples_per_second": 5.565, |
|
"eval_steps_per_second": 0.701, |
|
"step": 299 |
|
}, |
|
{ |
|
"epoch": 0.48764682057513165, |
|
"grad_norm": 1.7522519826889038, |
|
"learning_rate": 0.0001609131054601416, |
|
"loss": 0.5704, |
|
"step": 301 |
|
}, |
|
{ |
|
"epoch": 0.49898744430943703, |
|
"grad_norm": 1.4569737911224365, |
|
"learning_rate": 0.00015546017141067432, |
|
"loss": 0.5441, |
|
"step": 308 |
|
}, |
|
{ |
|
"epoch": 0.5054678007290401, |
|
"eval_loss": 0.5658866167068481, |
|
"eval_runtime": 100.9717, |
|
"eval_samples_per_second": 4.952, |
|
"eval_steps_per_second": 0.624, |
|
"step": 312 |
|
}, |
|
{ |
|
"epoch": 0.5103280680437424, |
|
"grad_norm": 1.8204305171966553, |
|
"learning_rate": 0.00015, |
|
"loss": 0.5583, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 0.5216686917780478, |
|
"grad_norm": 1.7424700260162354, |
|
"learning_rate": 0.0001445398285893257, |
|
"loss": 0.5468, |
|
"step": 322 |
|
}, |
|
{ |
|
"epoch": 0.5265289590927501, |
|
"eval_loss": 0.5584321022033691, |
|
"eval_runtime": 12.166, |
|
"eval_samples_per_second": 41.098, |
|
"eval_steps_per_second": 5.178, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 0.5330093155123532, |
|
"grad_norm": 1.5261849164962769, |
|
"learning_rate": 0.0001390868945398584, |
|
"loss": 0.547, |
|
"step": 329 |
|
}, |
|
{ |
|
"epoch": 0.5443499392466585, |
|
"grad_norm": 2.1022207736968994, |
|
"learning_rate": 0.00013364842561981014, |
|
"loss": 0.536, |
|
"step": 336 |
|
}, |
|
{ |
|
"epoch": 0.5475901174564601, |
|
"eval_loss": 0.554373025894165, |
|
"eval_runtime": 80.7911, |
|
"eval_samples_per_second": 6.189, |
|
"eval_steps_per_second": 0.78, |
|
"step": 338 |
|
}, |
|
{ |
|
"epoch": 0.555690562980964, |
|
"grad_norm": 1.375921607017517, |
|
"learning_rate": 0.00012823163042411807, |
|
"loss": 0.5271, |
|
"step": 343 |
|
}, |
|
{ |
|
"epoch": 0.5670311867152693, |
|
"grad_norm": 1.1903116703033447, |
|
"learning_rate": 0.0001228436888195789, |
|
"loss": 0.5277, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.56865127582017, |
|
"eval_loss": 0.5474019050598145, |
|
"eval_runtime": 91.4838, |
|
"eval_samples_per_second": 5.465, |
|
"eval_steps_per_second": 0.689, |
|
"step": 351 |
|
}, |
|
{ |
|
"epoch": 0.5783718104495748, |
|
"grad_norm": 1.5861282348632812, |
|
"learning_rate": 0.00011749174242806152, |
|
"loss": 0.5373, |
|
"step": 357 |
|
}, |
|
{ |
|
"epoch": 0.5897124341838801, |
|
"grad_norm": 1.3687732219696045, |
|
"learning_rate": 0.00011218288516041279, |
|
"loss": 0.5052, |
|
"step": 364 |
|
}, |
|
{ |
|
"epoch": 0.5897124341838801, |
|
"eval_loss": 0.5396790504455566, |
|
"eval_runtime": 12.1166, |
|
"eval_samples_per_second": 41.266, |
|
"eval_steps_per_second": 5.199, |
|
"step": 364 |
|
}, |
|
{ |
|
"epoch": 0.6010530579181855, |
|
"grad_norm": 1.3852468729019165, |
|
"learning_rate": 0.00010692415381360349, |
|
"loss": 0.5185, |
|
"step": 371 |
|
}, |
|
{ |
|
"epoch": 0.6107735925475901, |
|
"eval_loss": 0.5308696031570435, |
|
"eval_runtime": 82.1061, |
|
"eval_samples_per_second": 6.09, |
|
"eval_steps_per_second": 0.767, |
|
"step": 377 |
|
}, |
|
{ |
|
"epoch": 0.6123936816524909, |
|
"grad_norm": 1.3524726629257202, |
|
"learning_rate": 0.00010172251874357757, |
|
"loss": 0.5112, |
|
"step": 378 |
|
}, |
|
{ |
|
"epoch": 0.6237343053867963, |
|
"grad_norm": 1.8277941942214966, |
|
"learning_rate": 9.658487462616794e-05, |
|
"loss": 0.5161, |
|
"step": 385 |
|
}, |
|
{ |
|
"epoch": 0.6318347509113001, |
|
"eval_loss": 0.5261584520339966, |
|
"eval_runtime": 103.2902, |
|
"eval_samples_per_second": 4.841, |
|
"eval_steps_per_second": 0.61, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 0.6350749291211016, |
|
"grad_norm": 1.4200196266174316, |
|
"learning_rate": 9.151803131832493e-05, |
|
"loss": 0.5237, |
|
"step": 392 |
|
}, |
|
{ |
|
"epoch": 0.6464155528554071, |
|
"grad_norm": 1.3601250648498535, |
|
"learning_rate": 8.652870483177049e-05, |
|
"loss": 0.5056, |
|
"step": 399 |
|
}, |
|
{ |
|
"epoch": 0.6528959092750102, |
|
"eval_loss": 0.5227454304695129, |
|
"eval_runtime": 12.1817, |
|
"eval_samples_per_second": 41.045, |
|
"eval_steps_per_second": 5.172, |
|
"step": 403 |
|
}, |
|
{ |
|
"epoch": 0.6577561765897124, |
|
"grad_norm": 1.8182213306427002, |
|
"learning_rate": 8.162350843104291e-05, |
|
"loss": 0.5122, |
|
"step": 406 |
|
}, |
|
{ |
|
"epoch": 0.6690968003240179, |
|
"grad_norm": 1.909848928451538, |
|
"learning_rate": 7.680894386773072e-05, |
|
"loss": 0.5091, |
|
"step": 413 |
|
}, |
|
{ |
|
"epoch": 0.6739570676387201, |
|
"eval_loss": 0.5164161324501038, |
|
"eval_runtime": 82.8439, |
|
"eval_samples_per_second": 6.035, |
|
"eval_steps_per_second": 0.76, |
|
"step": 416 |
|
}, |
|
{ |
|
"epoch": 0.6804374240583232, |
|
"grad_norm": 1.4879406690597534, |
|
"learning_rate": 7.209139276251558e-05, |
|
"loss": 0.5023, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 0.6917780477926286, |
|
"grad_norm": 1.6504310369491577, |
|
"learning_rate": 6.747710814644589e-05, |
|
"loss": 0.492, |
|
"step": 427 |
|
}, |
|
{ |
|
"epoch": 0.6950182260024301, |
|
"eval_loss": 0.511544942855835, |
|
"eval_runtime": 83.9119, |
|
"eval_samples_per_second": 5.959, |
|
"eval_steps_per_second": 0.751, |
|
"step": 429 |
|
}, |
|
{ |
|
"epoch": 0.703118671526934, |
|
"grad_norm": 1.7625768184661865, |
|
"learning_rate": 6.297220617265435e-05, |
|
"loss": 0.484, |
|
"step": 434 |
|
}, |
|
{ |
|
"epoch": 0.7144592952612394, |
|
"grad_norm": 1.3635530471801758, |
|
"learning_rate": 5.858265800950438e-05, |
|
"loss": 0.4936, |
|
"step": 441 |
|
}, |
|
{ |
|
"epoch": 0.7160793843661402, |
|
"eval_loss": 0.506995677947998, |
|
"eval_runtime": 12.1546, |
|
"eval_samples_per_second": 41.137, |
|
"eval_steps_per_second": 5.183, |
|
"step": 442 |
|
}, |
|
{ |
|
"epoch": 0.7257999189955447, |
|
"grad_norm": 1.4940941333770752, |
|
"learning_rate": 5.4314281925911634e-05, |
|
"loss": 0.4918, |
|
"step": 448 |
|
}, |
|
{ |
|
"epoch": 0.7371405427298502, |
|
"grad_norm": 1.3953615427017212, |
|
"learning_rate": 5.0172735579330526e-05, |
|
"loss": 0.4818, |
|
"step": 455 |
|
}, |
|
{ |
|
"epoch": 0.7371405427298502, |
|
"eval_loss": 0.5005263686180115, |
|
"eval_runtime": 82.2428, |
|
"eval_samples_per_second": 6.08, |
|
"eval_steps_per_second": 0.766, |
|
"step": 455 |
|
}, |
|
{ |
|
"epoch": 0.7484811664641555, |
|
"grad_norm": 1.418354868888855, |
|
"learning_rate": 4.616350851662895e-05, |
|
"loss": 0.4762, |
|
"step": 462 |
|
}, |
|
{ |
|
"epoch": 0.7582017010935601, |
|
"eval_loss": 0.49856895208358765, |
|
"eval_runtime": 82.2573, |
|
"eval_samples_per_second": 6.078, |
|
"eval_steps_per_second": 0.766, |
|
"step": 468 |
|
}, |
|
{ |
|
"epoch": 0.759821790198461, |
|
"grad_norm": 1.4406681060791016, |
|
"learning_rate": 4.229191489779047e-05, |
|
"loss": 0.4979, |
|
"step": 469 |
|
}, |
|
{ |
|
"epoch": 0.7711624139327663, |
|
"grad_norm": 1.495307445526123, |
|
"learning_rate": 3.8563086452088506e-05, |
|
"loss": 0.4685, |
|
"step": 476 |
|
}, |
|
{ |
|
"epoch": 0.7792628594572701, |
|
"eval_loss": 0.4937572777271271, |
|
"eval_runtime": 12.1882, |
|
"eval_samples_per_second": 41.023, |
|
"eval_steps_per_second": 5.169, |
|
"step": 481 |
|
}, |
|
{ |
|
"epoch": 0.7825030376670717, |
|
"grad_norm": 1.4766854047775269, |
|
"learning_rate": 3.498196567606959e-05, |
|
"loss": 0.485, |
|
"step": 483 |
|
}, |
|
{ |
|
"epoch": 0.7938436614013771, |
|
"grad_norm": 1.3768912553787231, |
|
"learning_rate": 3.1553299282360966e-05, |
|
"loss": 0.4614, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 0.8003240178209802, |
|
"eval_loss": 0.49044370651245117, |
|
"eval_runtime": 82.6246, |
|
"eval_samples_per_second": 6.051, |
|
"eval_steps_per_second": 0.762, |
|
"step": 494 |
|
}, |
|
{ |
|
"epoch": 0.8051842851356824, |
|
"grad_norm": 1.2158797979354858, |
|
"learning_rate": 2.828163190798644e-05, |
|
"loss": 0.4779, |
|
"step": 497 |
|
}, |
|
{ |
|
"epoch": 0.8165249088699879, |
|
"grad_norm": 1.4826974868774414, |
|
"learning_rate": 2.5171300090530106e-05, |
|
"loss": 0.4942, |
|
"step": 504 |
|
}, |
|
{ |
|
"epoch": 0.8213851761846902, |
|
"eval_loss": 0.4870362877845764, |
|
"eval_runtime": 97.2746, |
|
"eval_samples_per_second": 5.14, |
|
"eval_steps_per_second": 0.648, |
|
"step": 507 |
|
}, |
|
{ |
|
"epoch": 0.8278655326042932, |
|
"grad_norm": 1.782825231552124, |
|
"learning_rate": 2.2226426520131734e-05, |
|
"loss": 0.4714, |
|
"step": 511 |
|
}, |
|
{ |
|
"epoch": 0.8392061563385986, |
|
"grad_norm": 1.2692053318023682, |
|
"learning_rate": 1.9450914574933725e-05, |
|
"loss": 0.4767, |
|
"step": 518 |
|
}, |
|
{ |
|
"epoch": 0.8424463345484001, |
|
"eval_loss": 0.48374512791633606, |
|
"eval_runtime": 12.151, |
|
"eval_samples_per_second": 41.149, |
|
"eval_steps_per_second": 5.185, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 0.850546780072904, |
|
"grad_norm": 1.1831132173538208, |
|
"learning_rate": 1.6848443147221828e-05, |
|
"loss": 0.4555, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 0.8618874038072094, |
|
"grad_norm": 1.2183641195297241, |
|
"learning_rate": 1.4422461767118233e-05, |
|
"loss": 0.4589, |
|
"step": 532 |
|
}, |
|
{ |
|
"epoch": 0.8635074929121102, |
|
"eval_loss": 0.48187872767448425, |
|
"eval_runtime": 85.4092, |
|
"eval_samples_per_second": 5.854, |
|
"eval_steps_per_second": 0.738, |
|
"step": 533 |
|
}, |
|
{ |
|
"epoch": 0.8732280275415147, |
|
"grad_norm": 1.3544665575027466, |
|
"learning_rate": 1.2176186030289936e-05, |
|
"loss": 0.4624, |
|
"step": 539 |
|
}, |
|
{ |
|
"epoch": 0.8845686512758202, |
|
"grad_norm": 1.4471445083618164, |
|
"learning_rate": 1.011259333573326e-05, |
|
"loss": 0.4806, |
|
"step": 546 |
|
}, |
|
{ |
|
"epoch": 0.8845686512758202, |
|
"eval_loss": 0.4796264171600342, |
|
"eval_runtime": 87.2314, |
|
"eval_samples_per_second": 5.732, |
|
"eval_steps_per_second": 0.722, |
|
"step": 546 |
|
}, |
|
{ |
|
"epoch": 0.8959092750101255, |
|
"grad_norm": 1.3122518062591553, |
|
"learning_rate": 8.234418939283866e-06, |
|
"loss": 0.4647, |
|
"step": 553 |
|
}, |
|
{ |
|
"epoch": 0.9056298096395302, |
|
"eval_loss": 0.4782348573207855, |
|
"eval_runtime": 12.1368, |
|
"eval_samples_per_second": 41.197, |
|
"eval_steps_per_second": 5.191, |
|
"step": 559 |
|
}, |
|
{ |
|
"epoch": 0.907249898744431, |
|
"grad_norm": 1.2790122032165527, |
|
"learning_rate": 6.544152328083152e-06, |
|
"loss": 0.4736, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 0.9185905224787363, |
|
"grad_norm": 1.3850630521774292, |
|
"learning_rate": 5.044033920806933e-06, |
|
"loss": 0.461, |
|
"step": 567 |
|
}, |
|
{ |
|
"epoch": 0.9266909680032401, |
|
"eval_loss": 0.4772534668445587, |
|
"eval_runtime": 55.6609, |
|
"eval_samples_per_second": 8.983, |
|
"eval_steps_per_second": 1.132, |
|
"step": 572 |
|
}, |
|
{ |
|
"epoch": 0.9299311462130417, |
|
"grad_norm": 1.3517329692840576, |
|
"learning_rate": 3.7360520980297514e-06, |
|
"loss": 0.4651, |
|
"step": 574 |
|
}, |
|
{ |
|
"epoch": 0.9412717699473471, |
|
"grad_norm": 1.5328634977340698, |
|
"learning_rate": 2.6219405666614402e-06, |
|
"loss": 0.4718, |
|
"step": 581 |
|
}, |
|
{ |
|
"epoch": 0.9477521263669502, |
|
"eval_loss": 0.476561576128006, |
|
"eval_runtime": 52.8481, |
|
"eval_samples_per_second": 9.461, |
|
"eval_steps_per_second": 1.192, |
|
"step": 585 |
|
}, |
|
{ |
|
"epoch": 0.9526123936816525, |
|
"grad_norm": 1.2334227561950684, |
|
"learning_rate": 1.7031760619491353e-06, |
|
"loss": 0.4506, |
|
"step": 588 |
|
}, |
|
{ |
|
"epoch": 0.9639530174159578, |
|
"grad_norm": 1.2170372009277344, |
|
"learning_rate": 9.809763900905875e-07, |
|
"loss": 0.4684, |
|
"step": 595 |
|
}, |
|
{ |
|
"epoch": 0.9688132847306602, |
|
"eval_loss": 0.47612592577934265, |
|
"eval_runtime": 12.1852, |
|
"eval_samples_per_second": 41.033, |
|
"eval_steps_per_second": 5.17, |
|
"step": 598 |
|
}, |
|
{ |
|
"epoch": 0.9752936411502633, |
|
"grad_norm": 1.179705023765564, |
|
"learning_rate": 4.562988140535073e-07, |
|
"loss": 0.4611, |
|
"step": 602 |
|
}, |
|
{ |
|
"epoch": 0.9866342648845686, |
|
"grad_norm": 1.2852600812911987, |
|
"learning_rate": 1.298387847403437e-07, |
|
"loss": 0.4716, |
|
"step": 609 |
|
}, |
|
{ |
|
"epoch": 0.9898744430943702, |
|
"eval_loss": 0.4760262072086334, |
|
"eval_runtime": 64.7528, |
|
"eval_samples_per_second": 7.722, |
|
"eval_steps_per_second": 0.973, |
|
"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": 7.394413366880502e+17, |
|
"train_batch_size": 8, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|