|
{ |
|
"best_metric": 4.443276405334473, |
|
"best_model_checkpoint": "/home/datta0/models/lora_final/Mistral-7B-v0.3_metamath_reverse/checkpoint-572", |
|
"epoch": 0.9995949777237748, |
|
"eval_steps": 13, |
|
"global_step": 617, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0016200891049007696, |
|
"grad_norm": 31.96648597717285, |
|
"learning_rate": 2.3076923076923076e-05, |
|
"loss": 0.9728, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.011340623734305387, |
|
"grad_norm": 25.136903762817383, |
|
"learning_rate": 0.00016153846153846153, |
|
"loss": 0.7436, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 0.021061158363710003, |
|
"eval_loss": 7.4054484367370605, |
|
"eval_runtime": 12.8928, |
|
"eval_samples_per_second": 38.781, |
|
"eval_steps_per_second": 4.886, |
|
"step": 13 |
|
}, |
|
{ |
|
"epoch": 0.022681247468610773, |
|
"grad_norm": 6121.2451171875, |
|
"learning_rate": 0.0002999979709808197, |
|
"loss": 1.928, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 0.03402187120291616, |
|
"grad_norm": 30.988468170166016, |
|
"learning_rate": 0.0002998701612152596, |
|
"loss": 9.1068, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 0.042122316727420006, |
|
"eval_loss": 6.97995662689209, |
|
"eval_runtime": 88.3905, |
|
"eval_samples_per_second": 5.657, |
|
"eval_steps_per_second": 0.713, |
|
"step": 26 |
|
}, |
|
{ |
|
"epoch": 0.04536249493722155, |
|
"grad_norm": 5.160550117492676, |
|
"learning_rate": 0.0002995437011859465, |
|
"loss": 7.0452, |
|
"step": 28 |
|
}, |
|
{ |
|
"epoch": 0.056703118671526935, |
|
"grad_norm": 4.622859954833984, |
|
"learning_rate": 0.00029901902360990936, |
|
"loss": 6.6988, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 0.06318347509113001, |
|
"eval_loss": 6.427122116088867, |
|
"eval_runtime": 94.0313, |
|
"eval_samples_per_second": 5.317, |
|
"eval_steps_per_second": 0.67, |
|
"step": 39 |
|
}, |
|
{ |
|
"epoch": 0.06804374240583232, |
|
"grad_norm": 419.7519226074219, |
|
"learning_rate": 0.00029829682393805085, |
|
"loss": 6.4471, |
|
"step": 42 |
|
}, |
|
{ |
|
"epoch": 0.0793843661401377, |
|
"grad_norm": 3.067017078399658, |
|
"learning_rate": 0.0002973780594333385, |
|
"loss": 6.4684, |
|
"step": 49 |
|
}, |
|
{ |
|
"epoch": 0.08424463345484001, |
|
"eval_loss": 6.289278507232666, |
|
"eval_runtime": 12.7505, |
|
"eval_samples_per_second": 39.214, |
|
"eval_steps_per_second": 4.941, |
|
"step": 52 |
|
}, |
|
{ |
|
"epoch": 0.0907249898744431, |
|
"grad_norm": 14.996749877929688, |
|
"learning_rate": 0.00029626394790197025, |
|
"loss": 6.2403, |
|
"step": 56 |
|
}, |
|
{ |
|
"epoch": 0.10206561360874848, |
|
"grad_norm": 13.672094345092773, |
|
"learning_rate": 0.00029495596607919305, |
|
"loss": 6.1245, |
|
"step": 63 |
|
}, |
|
{ |
|
"epoch": 0.10530579181855002, |
|
"eval_loss": 6.124541759490967, |
|
"eval_runtime": 92.7829, |
|
"eval_samples_per_second": 5.389, |
|
"eval_steps_per_second": 0.679, |
|
"step": 65 |
|
}, |
|
{ |
|
"epoch": 0.11340623734305387, |
|
"grad_norm": 8.664388656616211, |
|
"learning_rate": 0.00029345584767191685, |
|
"loss": 6.0178, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.12474686107735926, |
|
"grad_norm": 11.606036186218262, |
|
"learning_rate": 0.0002917655810607161, |
|
"loss": 5.9117, |
|
"step": 77 |
|
}, |
|
{ |
|
"epoch": 0.12636695018226002, |
|
"eval_loss": 5.8770036697387695, |
|
"eval_runtime": 84.5458, |
|
"eval_samples_per_second": 5.914, |
|
"eval_steps_per_second": 0.745, |
|
"step": 78 |
|
}, |
|
{ |
|
"epoch": 0.13608748481166463, |
|
"grad_norm": 18.062744140625, |
|
"learning_rate": 0.0002898874066642667, |
|
"loss": 5.8276, |
|
"step": 84 |
|
}, |
|
{ |
|
"epoch": 0.14742810854597002, |
|
"grad_norm": 28.24692153930664, |
|
"learning_rate": 0.00028782381396971003, |
|
"loss": 5.8448, |
|
"step": 91 |
|
}, |
|
{ |
|
"epoch": 0.14742810854597002, |
|
"eval_loss": 5.783429145812988, |
|
"eval_runtime": 12.9352, |
|
"eval_samples_per_second": 38.654, |
|
"eval_steps_per_second": 4.87, |
|
"step": 91 |
|
}, |
|
{ |
|
"epoch": 0.1587687322802754, |
|
"grad_norm": 25.64031982421875, |
|
"learning_rate": 0.00028557753823288173, |
|
"loss": 5.742, |
|
"step": 98 |
|
}, |
|
{ |
|
"epoch": 0.16848926690968002, |
|
"eval_loss": 5.894050598144531, |
|
"eval_runtime": 88.9837, |
|
"eval_samples_per_second": 5.619, |
|
"eval_steps_per_second": 0.708, |
|
"step": 104 |
|
}, |
|
{ |
|
"epoch": 0.1701093560145808, |
|
"grad_norm": 19.707569122314453, |
|
"learning_rate": 0.0002831515568527781, |
|
"loss": 5.7521, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 0.1814499797488862, |
|
"grad_norm": 17.04213523864746, |
|
"learning_rate": 0.00028054908542506627, |
|
"loss": 5.6054, |
|
"step": 112 |
|
}, |
|
{ |
|
"epoch": 0.18955042527339003, |
|
"eval_loss": 6.097167015075684, |
|
"eval_runtime": 84.3979, |
|
"eval_samples_per_second": 5.924, |
|
"eval_steps_per_second": 0.746, |
|
"step": 117 |
|
}, |
|
{ |
|
"epoch": 0.19279060348319157, |
|
"grad_norm": 92.46790313720703, |
|
"learning_rate": 0.00027777357347986823, |
|
"loss": 5.801, |
|
"step": 119 |
|
}, |
|
{ |
|
"epoch": 0.20413122721749696, |
|
"grad_norm": 24.985816955566406, |
|
"learning_rate": 0.00027482869990946986, |
|
"loss": 5.6465, |
|
"step": 126 |
|
}, |
|
{ |
|
"epoch": 0.21061158363710003, |
|
"eval_loss": 5.480843544006348, |
|
"eval_runtime": 12.9876, |
|
"eval_samples_per_second": 38.498, |
|
"eval_steps_per_second": 4.851, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.21547185095180235, |
|
"grad_norm": 18.881345748901367, |
|
"learning_rate": 0.0002717183680920135, |
|
"loss": 5.5278, |
|
"step": 133 |
|
}, |
|
{ |
|
"epoch": 0.22681247468610774, |
|
"grad_norm": 21.324201583862305, |
|
"learning_rate": 0.00026844670071763906, |
|
"loss": 5.5659, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.23167274200081003, |
|
"eval_loss": 5.537073135375977, |
|
"eval_runtime": 85.8424, |
|
"eval_samples_per_second": 5.825, |
|
"eval_steps_per_second": 0.734, |
|
"step": 143 |
|
}, |
|
{ |
|
"epoch": 0.23815309842041313, |
|
"grad_norm": 26.40949249267578, |
|
"learning_rate": 0.00026501803432393037, |
|
"loss": 5.5108, |
|
"step": 147 |
|
}, |
|
{ |
|
"epoch": 0.24949372215471852, |
|
"grad_norm": 55.612056732177734, |
|
"learning_rate": 0.00026143691354791145, |
|
"loss": 5.4175, |
|
"step": 154 |
|
}, |
|
{ |
|
"epoch": 0.25273390036452004, |
|
"eval_loss": 5.568782329559326, |
|
"eval_runtime": 94.8092, |
|
"eval_samples_per_second": 5.274, |
|
"eval_steps_per_second": 0.664, |
|
"step": 156 |
|
}, |
|
{ |
|
"epoch": 0.2608343458890239, |
|
"grad_norm": 14.32459545135498, |
|
"learning_rate": 0.00025770808510220956, |
|
"loss": 5.3415, |
|
"step": 161 |
|
}, |
|
{ |
|
"epoch": 0.27217496962332927, |
|
"grad_norm": 38.13811111450195, |
|
"learning_rate": 0.00025383649148337105, |
|
"loss": 5.3148, |
|
"step": 168 |
|
}, |
|
{ |
|
"epoch": 0.27379505872823007, |
|
"eval_loss": 5.364619731903076, |
|
"eval_runtime": 12.9741, |
|
"eval_samples_per_second": 38.538, |
|
"eval_steps_per_second": 4.856, |
|
"step": 169 |
|
}, |
|
{ |
|
"epoch": 0.28351559335763465, |
|
"grad_norm": 17.65839385986328, |
|
"learning_rate": 0.0002498272644206695, |
|
"loss": 5.2599, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 0.29485621709194004, |
|
"grad_norm": 9.63137435913086, |
|
"learning_rate": 0.0002456857180740884, |
|
"loss": 5.2376, |
|
"step": 182 |
|
}, |
|
{ |
|
"epoch": 0.29485621709194004, |
|
"eval_loss": 5.205206871032715, |
|
"eval_runtime": 94.3725, |
|
"eval_samples_per_second": 5.298, |
|
"eval_steps_per_second": 0.668, |
|
"step": 182 |
|
}, |
|
{ |
|
"epoch": 0.30619684082624543, |
|
"grad_norm": 8.901846885681152, |
|
"learning_rate": 0.0002414173419904956, |
|
"loss": 5.2313, |
|
"step": 189 |
|
}, |
|
{ |
|
"epoch": 0.3159173754556501, |
|
"eval_loss": 5.147277355194092, |
|
"eval_runtime": 90.8266, |
|
"eval_samples_per_second": 5.505, |
|
"eval_steps_per_second": 0.694, |
|
"step": 195 |
|
}, |
|
{ |
|
"epoch": 0.3175374645605508, |
|
"grad_norm": 16.01649284362793, |
|
"learning_rate": 0.00023702779382734566, |
|
"loss": 5.1989, |
|
"step": 196 |
|
}, |
|
{ |
|
"epoch": 0.3288780882948562, |
|
"grad_norm": 35.22917556762695, |
|
"learning_rate": 0.0002325228918535541, |
|
"loss": 5.1381, |
|
"step": 203 |
|
}, |
|
{ |
|
"epoch": 0.33697853381936005, |
|
"eval_loss": 5.247087001800537, |
|
"eval_runtime": 13.0295, |
|
"eval_samples_per_second": 38.375, |
|
"eval_steps_per_second": 4.835, |
|
"step": 208 |
|
}, |
|
{ |
|
"epoch": 0.3402187120291616, |
|
"grad_norm": 18.41366958618164, |
|
"learning_rate": 0.00022790860723748442, |
|
"loss": 5.2052, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.351559335763467, |
|
"grad_norm": 36.26469802856445, |
|
"learning_rate": 0.00022319105613226921, |
|
"loss": 5.0545, |
|
"step": 217 |
|
}, |
|
{ |
|
"epoch": 0.3580396921830701, |
|
"eval_loss": 5.057940483093262, |
|
"eval_runtime": 81.209, |
|
"eval_samples_per_second": 6.157, |
|
"eval_steps_per_second": 0.776, |
|
"step": 221 |
|
}, |
|
{ |
|
"epoch": 0.3628999594977724, |
|
"grad_norm": 35.501441955566406, |
|
"learning_rate": 0.00021837649156895706, |
|
"loss": 5.0624, |
|
"step": 224 |
|
}, |
|
{ |
|
"epoch": 0.37424058323207776, |
|
"grad_norm": 25.91912078857422, |
|
"learning_rate": 0.00021347129516822945, |
|
"loss": 5.0218, |
|
"step": 231 |
|
}, |
|
{ |
|
"epoch": 0.37910085054678005, |
|
"eval_loss": 5.043420314788818, |
|
"eval_runtime": 81.4809, |
|
"eval_samples_per_second": 6.136, |
|
"eval_steps_per_second": 0.773, |
|
"step": 234 |
|
}, |
|
{ |
|
"epoch": 0.38558120696638315, |
|
"grad_norm": 76.39311981201172, |
|
"learning_rate": 0.00020848196868167505, |
|
"loss": 5.0395, |
|
"step": 238 |
|
}, |
|
{ |
|
"epoch": 0.39692183070068854, |
|
"grad_norm": 55.80715560913086, |
|
"learning_rate": 0.000203415125373832, |
|
"loss": 5.1901, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 0.4001620089104901, |
|
"eval_loss": 5.186158180236816, |
|
"eval_runtime": 13.0415, |
|
"eval_samples_per_second": 38.339, |
|
"eval_steps_per_second": 4.831, |
|
"step": 247 |
|
}, |
|
{ |
|
"epoch": 0.4082624544349939, |
|
"grad_norm": 44.12828063964844, |
|
"learning_rate": 0.00019827748125642242, |
|
"loss": 5.2192, |
|
"step": 252 |
|
}, |
|
{ |
|
"epoch": 0.4196030781692993, |
|
"grad_norm": 20.139680862426758, |
|
"learning_rate": 0.0001930758461863965, |
|
"loss": 5.0809, |
|
"step": 259 |
|
}, |
|
{ |
|
"epoch": 0.42122316727420006, |
|
"eval_loss": 5.010337829589844, |
|
"eval_runtime": 89.8841, |
|
"eval_samples_per_second": 5.563, |
|
"eval_steps_per_second": 0.701, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.4309437019036047, |
|
"grad_norm": 22.906221389770508, |
|
"learning_rate": 0.0001878171148395872, |
|
"loss": 5.0478, |
|
"step": 266 |
|
}, |
|
{ |
|
"epoch": 0.4422843256379101, |
|
"grad_norm": 30.48619270324707, |
|
"learning_rate": 0.00018250825757193848, |
|
"loss": 5.0357, |
|
"step": 273 |
|
}, |
|
{ |
|
"epoch": 0.4422843256379101, |
|
"eval_loss": 5.04883337020874, |
|
"eval_runtime": 93.6727, |
|
"eval_samples_per_second": 5.338, |
|
"eval_steps_per_second": 0.673, |
|
"step": 273 |
|
}, |
|
{ |
|
"epoch": 0.4536249493722155, |
|
"grad_norm": 69.97283172607422, |
|
"learning_rate": 0.0001771563111804211, |
|
"loss": 5.0375, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.46334548400162007, |
|
"eval_loss": 5.002610206604004, |
|
"eval_runtime": 13.0501, |
|
"eval_samples_per_second": 38.314, |
|
"eval_steps_per_second": 4.828, |
|
"step": 286 |
|
}, |
|
{ |
|
"epoch": 0.46496557310652087, |
|
"grad_norm": 69.82508850097656, |
|
"learning_rate": 0.0001717683695758819, |
|
"loss": 4.9778, |
|
"step": 287 |
|
}, |
|
{ |
|
"epoch": 0.47630619684082626, |
|
"grad_norm": 95.04804229736328, |
|
"learning_rate": 0.00016635157438018983, |
|
"loss": 5.0348, |
|
"step": 294 |
|
}, |
|
{ |
|
"epoch": 0.4844066423653301, |
|
"eval_loss": 5.008110523223877, |
|
"eval_runtime": 86.595, |
|
"eval_samples_per_second": 5.774, |
|
"eval_steps_per_second": 0.728, |
|
"step": 299 |
|
}, |
|
{ |
|
"epoch": 0.48764682057513165, |
|
"grad_norm": 62.681434631347656, |
|
"learning_rate": 0.0001609131054601416, |
|
"loss": 5.0078, |
|
"step": 301 |
|
}, |
|
{ |
|
"epoch": 0.49898744430943703, |
|
"grad_norm": 58.10196304321289, |
|
"learning_rate": 0.00015546017141067432, |
|
"loss": 4.8927, |
|
"step": 308 |
|
}, |
|
{ |
|
"epoch": 0.5054678007290401, |
|
"eval_loss": 4.8912458419799805, |
|
"eval_runtime": 77.9462, |
|
"eval_samples_per_second": 6.415, |
|
"eval_steps_per_second": 0.808, |
|
"step": 312 |
|
}, |
|
{ |
|
"epoch": 0.5103280680437424, |
|
"grad_norm": 44.00553512573242, |
|
"learning_rate": 0.00015, |
|
"loss": 4.9021, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 0.5216686917780478, |
|
"grad_norm": 39.55318832397461, |
|
"learning_rate": 0.0001445398285893257, |
|
"loss": 4.878, |
|
"step": 322 |
|
}, |
|
{ |
|
"epoch": 0.5265289590927501, |
|
"eval_loss": 4.866469860076904, |
|
"eval_runtime": 13.1009, |
|
"eval_samples_per_second": 38.165, |
|
"eval_steps_per_second": 4.809, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 0.5330093155123532, |
|
"grad_norm": 22.130632400512695, |
|
"learning_rate": 0.0001390868945398584, |
|
"loss": 4.8018, |
|
"step": 329 |
|
}, |
|
{ |
|
"epoch": 0.5443499392466585, |
|
"grad_norm": 46.44150161743164, |
|
"learning_rate": 0.00013364842561981014, |
|
"loss": 4.8092, |
|
"step": 336 |
|
}, |
|
{ |
|
"epoch": 0.5475901174564601, |
|
"eval_loss": 4.84024715423584, |
|
"eval_runtime": 94.4317, |
|
"eval_samples_per_second": 5.295, |
|
"eval_steps_per_second": 0.667, |
|
"step": 338 |
|
}, |
|
{ |
|
"epoch": 0.555690562980964, |
|
"grad_norm": 28.048892974853516, |
|
"learning_rate": 0.00012823163042411807, |
|
"loss": 4.7609, |
|
"step": 343 |
|
}, |
|
{ |
|
"epoch": 0.5670311867152693, |
|
"grad_norm": 49.06658935546875, |
|
"learning_rate": 0.0001228436888195789, |
|
"loss": 4.8342, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.56865127582017, |
|
"eval_loss": 4.768944263458252, |
|
"eval_runtime": 95.9831, |
|
"eval_samples_per_second": 5.209, |
|
"eval_steps_per_second": 0.656, |
|
"step": 351 |
|
}, |
|
{ |
|
"epoch": 0.5783718104495748, |
|
"grad_norm": 51.22993469238281, |
|
"learning_rate": 0.00011749174242806152, |
|
"loss": 4.7839, |
|
"step": 357 |
|
}, |
|
{ |
|
"epoch": 0.5897124341838801, |
|
"grad_norm": 18.450580596923828, |
|
"learning_rate": 0.00011218288516041279, |
|
"loss": 4.7834, |
|
"step": 364 |
|
}, |
|
{ |
|
"epoch": 0.5897124341838801, |
|
"eval_loss": 4.78420352935791, |
|
"eval_runtime": 13.1163, |
|
"eval_samples_per_second": 38.121, |
|
"eval_steps_per_second": 4.803, |
|
"step": 364 |
|
}, |
|
{ |
|
"epoch": 0.6010530579181855, |
|
"grad_norm": 20.635562896728516, |
|
"learning_rate": 0.00010692415381360349, |
|
"loss": 4.7428, |
|
"step": 371 |
|
}, |
|
{ |
|
"epoch": 0.6107735925475901, |
|
"eval_loss": 4.739642143249512, |
|
"eval_runtime": 85.1259, |
|
"eval_samples_per_second": 5.874, |
|
"eval_steps_per_second": 0.74, |
|
"step": 377 |
|
}, |
|
{ |
|
"epoch": 0.6123936816524909, |
|
"grad_norm": 23.58678436279297, |
|
"learning_rate": 0.00010172251874357757, |
|
"loss": 4.7561, |
|
"step": 378 |
|
}, |
|
{ |
|
"epoch": 0.6237343053867963, |
|
"grad_norm": 15.554181098937988, |
|
"learning_rate": 9.658487462616794e-05, |
|
"loss": 4.7318, |
|
"step": 385 |
|
}, |
|
{ |
|
"epoch": 0.6318347509113001, |
|
"eval_loss": 4.698749542236328, |
|
"eval_runtime": 89.829, |
|
"eval_samples_per_second": 5.566, |
|
"eval_steps_per_second": 0.701, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 0.6350749291211016, |
|
"grad_norm": 11.260749816894531, |
|
"learning_rate": 9.151803131832493e-05, |
|
"loss": 4.7366, |
|
"step": 392 |
|
}, |
|
{ |
|
"epoch": 0.6464155528554071, |
|
"grad_norm": 23.6870059967041, |
|
"learning_rate": 8.652870483177049e-05, |
|
"loss": 4.6442, |
|
"step": 399 |
|
}, |
|
{ |
|
"epoch": 0.6528959092750102, |
|
"eval_loss": 4.6853532791137695, |
|
"eval_runtime": 13.1, |
|
"eval_samples_per_second": 38.168, |
|
"eval_steps_per_second": 4.809, |
|
"step": 403 |
|
}, |
|
{ |
|
"epoch": 0.6577561765897124, |
|
"grad_norm": 27.518665313720703, |
|
"learning_rate": 8.162350843104291e-05, |
|
"loss": 4.6589, |
|
"step": 406 |
|
}, |
|
{ |
|
"epoch": 0.6690968003240179, |
|
"grad_norm": 40.192344665527344, |
|
"learning_rate": 7.680894386773072e-05, |
|
"loss": 4.6454, |
|
"step": 413 |
|
}, |
|
{ |
|
"epoch": 0.6739570676387201, |
|
"eval_loss": 4.691702365875244, |
|
"eval_runtime": 89.1902, |
|
"eval_samples_per_second": 5.606, |
|
"eval_steps_per_second": 0.706, |
|
"step": 416 |
|
}, |
|
{ |
|
"epoch": 0.6804374240583232, |
|
"grad_norm": 23.988582611083984, |
|
"learning_rate": 7.209139276251558e-05, |
|
"loss": 4.6533, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 0.6917780477926286, |
|
"grad_norm": 30.93518829345703, |
|
"learning_rate": 6.747710814644589e-05, |
|
"loss": 4.7075, |
|
"step": 427 |
|
}, |
|
{ |
|
"epoch": 0.6950182260024301, |
|
"eval_loss": 4.641927719116211, |
|
"eval_runtime": 91.2485, |
|
"eval_samples_per_second": 5.48, |
|
"eval_steps_per_second": 0.69, |
|
"step": 429 |
|
}, |
|
{ |
|
"epoch": 0.703118671526934, |
|
"grad_norm": 28.248205184936523, |
|
"learning_rate": 6.297220617265435e-05, |
|
"loss": 4.608, |
|
"step": 434 |
|
}, |
|
{ |
|
"epoch": 0.7144592952612394, |
|
"grad_norm": 16.064809799194336, |
|
"learning_rate": 5.858265800950438e-05, |
|
"loss": 4.6744, |
|
"step": 441 |
|
}, |
|
{ |
|
"epoch": 0.7160793843661402, |
|
"eval_loss": 4.58261251449585, |
|
"eval_runtime": 13.1224, |
|
"eval_samples_per_second": 38.103, |
|
"eval_steps_per_second": 4.801, |
|
"step": 442 |
|
}, |
|
{ |
|
"epoch": 0.7257999189955447, |
|
"grad_norm": 12.395214080810547, |
|
"learning_rate": 5.4314281925911634e-05, |
|
"loss": 4.5883, |
|
"step": 448 |
|
}, |
|
{ |
|
"epoch": 0.7371405427298502, |
|
"grad_norm": 15.379476547241211, |
|
"learning_rate": 5.0172735579330526e-05, |
|
"loss": 4.5861, |
|
"step": 455 |
|
}, |
|
{ |
|
"epoch": 0.7371405427298502, |
|
"eval_loss": 4.579273700714111, |
|
"eval_runtime": 84.8776, |
|
"eval_samples_per_second": 5.891, |
|
"eval_steps_per_second": 0.742, |
|
"step": 455 |
|
}, |
|
{ |
|
"epoch": 0.7484811664641555, |
|
"grad_norm": 10.24847412109375, |
|
"learning_rate": 4.616350851662895e-05, |
|
"loss": 4.5707, |
|
"step": 462 |
|
}, |
|
{ |
|
"epoch": 0.7582017010935601, |
|
"eval_loss": 4.59439754486084, |
|
"eval_runtime": 92.5346, |
|
"eval_samples_per_second": 5.403, |
|
"eval_steps_per_second": 0.681, |
|
"step": 468 |
|
}, |
|
{ |
|
"epoch": 0.759821790198461, |
|
"grad_norm": 20.303340911865234, |
|
"learning_rate": 4.229191489779047e-05, |
|
"loss": 4.5872, |
|
"step": 469 |
|
}, |
|
{ |
|
"epoch": 0.7711624139327663, |
|
"grad_norm": 45.71290969848633, |
|
"learning_rate": 3.8563086452088506e-05, |
|
"loss": 4.5675, |
|
"step": 476 |
|
}, |
|
{ |
|
"epoch": 0.7792628594572701, |
|
"eval_loss": 4.561083793640137, |
|
"eval_runtime": 13.1075, |
|
"eval_samples_per_second": 38.146, |
|
"eval_steps_per_second": 4.806, |
|
"step": 481 |
|
}, |
|
{ |
|
"epoch": 0.7825030376670717, |
|
"grad_norm": 39.32904052734375, |
|
"learning_rate": 3.498196567606959e-05, |
|
"loss": 4.5437, |
|
"step": 483 |
|
}, |
|
{ |
|
"epoch": 0.7938436614013771, |
|
"grad_norm": 18.427520751953125, |
|
"learning_rate": 3.1553299282360966e-05, |
|
"loss": 4.5286, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 0.8003240178209802, |
|
"eval_loss": 4.521579265594482, |
|
"eval_runtime": 92.4799, |
|
"eval_samples_per_second": 5.407, |
|
"eval_steps_per_second": 0.681, |
|
"step": 494 |
|
}, |
|
{ |
|
"epoch": 0.8051842851356824, |
|
"grad_norm": 12.541765213012695, |
|
"learning_rate": 2.828163190798644e-05, |
|
"loss": 4.48, |
|
"step": 497 |
|
}, |
|
{ |
|
"epoch": 0.8165249088699879, |
|
"grad_norm": 14.985490798950195, |
|
"learning_rate": 2.5171300090530106e-05, |
|
"loss": 4.5302, |
|
"step": 504 |
|
}, |
|
{ |
|
"epoch": 0.8213851761846902, |
|
"eval_loss": 4.5139875411987305, |
|
"eval_runtime": 88.2227, |
|
"eval_samples_per_second": 5.667, |
|
"eval_steps_per_second": 0.714, |
|
"step": 507 |
|
}, |
|
{ |
|
"epoch": 0.8278655326042932, |
|
"grad_norm": 28.087665557861328, |
|
"learning_rate": 2.2226426520131734e-05, |
|
"loss": 4.464, |
|
"step": 511 |
|
}, |
|
{ |
|
"epoch": 0.8392061563385986, |
|
"grad_norm": 17.76738739013672, |
|
"learning_rate": 1.9450914574933725e-05, |
|
"loss": 4.5191, |
|
"step": 518 |
|
}, |
|
{ |
|
"epoch": 0.8424463345484001, |
|
"eval_loss": 4.508421897888184, |
|
"eval_runtime": 13.1321, |
|
"eval_samples_per_second": 38.075, |
|
"eval_steps_per_second": 4.797, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 0.850546780072904, |
|
"grad_norm": 28.11948585510254, |
|
"learning_rate": 1.6848443147221828e-05, |
|
"loss": 4.4829, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 0.8618874038072094, |
|
"grad_norm": 29.396305084228516, |
|
"learning_rate": 1.4422461767118233e-05, |
|
"loss": 4.5023, |
|
"step": 532 |
|
}, |
|
{ |
|
"epoch": 0.8635074929121102, |
|
"eval_loss": 4.487783432006836, |
|
"eval_runtime": 91.7711, |
|
"eval_samples_per_second": 5.448, |
|
"eval_steps_per_second": 0.686, |
|
"step": 533 |
|
}, |
|
{ |
|
"epoch": 0.8732280275415147, |
|
"grad_norm": 16.643442153930664, |
|
"learning_rate": 1.2176186030289936e-05, |
|
"loss": 4.4684, |
|
"step": 539 |
|
}, |
|
{ |
|
"epoch": 0.8845686512758202, |
|
"grad_norm": 6.223151683807373, |
|
"learning_rate": 1.011259333573326e-05, |
|
"loss": 4.4661, |
|
"step": 546 |
|
}, |
|
{ |
|
"epoch": 0.8845686512758202, |
|
"eval_loss": 4.459321022033691, |
|
"eval_runtime": 80.8756, |
|
"eval_samples_per_second": 6.182, |
|
"eval_steps_per_second": 0.779, |
|
"step": 546 |
|
}, |
|
{ |
|
"epoch": 0.8959092750101255, |
|
"grad_norm": 11.320446968078613, |
|
"learning_rate": 8.234418939283866e-06, |
|
"loss": 4.4942, |
|
"step": 553 |
|
}, |
|
{ |
|
"epoch": 0.9056298096395302, |
|
"eval_loss": 4.465951919555664, |
|
"eval_runtime": 13.1729, |
|
"eval_samples_per_second": 37.957, |
|
"eval_steps_per_second": 4.783, |
|
"step": 559 |
|
}, |
|
{ |
|
"epoch": 0.907249898744431, |
|
"grad_norm": 14.810989379882812, |
|
"learning_rate": 6.544152328083152e-06, |
|
"loss": 4.4924, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 0.9185905224787363, |
|
"grad_norm": 7.798120975494385, |
|
"learning_rate": 5.044033920806933e-06, |
|
"loss": 4.4691, |
|
"step": 567 |
|
}, |
|
{ |
|
"epoch": 0.9266909680032401, |
|
"eval_loss": 4.443276405334473, |
|
"eval_runtime": 87.1678, |
|
"eval_samples_per_second": 5.736, |
|
"eval_steps_per_second": 0.723, |
|
"step": 572 |
|
}, |
|
{ |
|
"epoch": 0.9299311462130417, |
|
"grad_norm": 8.354825973510742, |
|
"learning_rate": 3.7360520980297514e-06, |
|
"loss": 4.4501, |
|
"step": 574 |
|
}, |
|
{ |
|
"epoch": 0.9412717699473471, |
|
"grad_norm": 12.046449661254883, |
|
"learning_rate": 2.6219405666614402e-06, |
|
"loss": 4.4153, |
|
"step": 581 |
|
}, |
|
{ |
|
"epoch": 0.9477521263669502, |
|
"eval_loss": 4.453136920928955, |
|
"eval_runtime": 96.3934, |
|
"eval_samples_per_second": 5.187, |
|
"eval_steps_per_second": 0.654, |
|
"step": 585 |
|
}, |
|
{ |
|
"epoch": 0.9526123936816525, |
|
"grad_norm": 10.31865406036377, |
|
"learning_rate": 1.7031760619491353e-06, |
|
"loss": 4.4468, |
|
"step": 588 |
|
}, |
|
{ |
|
"epoch": 0.9639530174159578, |
|
"grad_norm": 11.857120513916016, |
|
"learning_rate": 9.809763900905875e-07, |
|
"loss": 4.4609, |
|
"step": 595 |
|
}, |
|
{ |
|
"epoch": 0.9688132847306602, |
|
"eval_loss": 4.445005893707275, |
|
"eval_runtime": 13.1103, |
|
"eval_samples_per_second": 38.138, |
|
"eval_steps_per_second": 4.805, |
|
"step": 598 |
|
}, |
|
{ |
|
"epoch": 0.9752936411502633, |
|
"grad_norm": 7.414619445800781, |
|
"learning_rate": 4.562988140535073e-07, |
|
"loss": 4.4396, |
|
"step": 602 |
|
}, |
|
{ |
|
"epoch": 0.9866342648845686, |
|
"grad_norm": 7.68168830871582, |
|
"learning_rate": 1.298387847403437e-07, |
|
"loss": 4.4444, |
|
"step": 609 |
|
}, |
|
{ |
|
"epoch": 0.9898744430943702, |
|
"eval_loss": 4.447805881500244, |
|
"eval_runtime": 94.8799, |
|
"eval_samples_per_second": 5.27, |
|
"eval_steps_per_second": 0.664, |
|
"step": 611 |
|
}, |
|
{ |
|
"epoch": 0.9979748886188741, |
|
"grad_norm": 8.450008392333984, |
|
"learning_rate": 2.029019180288527e-09, |
|
"loss": 4.409, |
|
"step": 616 |
|
} |
|
], |
|
"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": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 8.001836583160381e+17, |
|
"train_batch_size": 8, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|