| { | |
| "best_metric": 0.18747980892658234, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Qwen2-7B_metamath_reverse/checkpoint-13", | |
| "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": 4.52362585067749, | |
| "learning_rate": 2.3076923076923076e-05, | |
| "loss": 0.2546, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011340623734305387, | |
| "grad_norm": 1.5261310338974, | |
| "learning_rate": 0.00016153846153846153, | |
| "loss": 0.1753, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.021061158363710003, | |
| "eval_loss": 0.18747980892658234, | |
| "eval_runtime": 12.4138, | |
| "eval_samples_per_second": 40.278, | |
| "eval_steps_per_second": 5.075, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.022681247468610773, | |
| "grad_norm": 2.1798665523529053, | |
| "learning_rate": 0.0002999979709808197, | |
| "loss": 0.1715, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03402187120291616, | |
| "grad_norm": 2.5627083778381348, | |
| "learning_rate": 0.0002998701612152596, | |
| "loss": 0.2036, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.042122316727420006, | |
| "eval_loss": 0.2523616552352905, | |
| "eval_runtime": 83.5071, | |
| "eval_samples_per_second": 5.988, | |
| "eval_steps_per_second": 0.754, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.04536249493722155, | |
| "grad_norm": 2.336559772491455, | |
| "learning_rate": 0.0002995437011859465, | |
| "loss": 0.2407, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.056703118671526935, | |
| "grad_norm": 2.8058784008026123, | |
| "learning_rate": 0.00029901902360990936, | |
| "loss": 0.2585, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.06318347509113001, | |
| "eval_loss": 0.2876286506652832, | |
| "eval_runtime": 74.7685, | |
| "eval_samples_per_second": 6.687, | |
| "eval_steps_per_second": 0.843, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.06804374240583232, | |
| "grad_norm": 2.3905534744262695, | |
| "learning_rate": 0.00029829682393805085, | |
| "loss": 0.2638, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0793843661401377, | |
| "grad_norm": 2.4771692752838135, | |
| "learning_rate": 0.0002973780594333385, | |
| "loss": 0.2848, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.08424463345484001, | |
| "eval_loss": 0.31458961963653564, | |
| "eval_runtime": 12.4219, | |
| "eval_samples_per_second": 40.252, | |
| "eval_steps_per_second": 5.072, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.0907249898744431, | |
| "grad_norm": 2.488586902618408, | |
| "learning_rate": 0.00029626394790197025, | |
| "loss": 0.3027, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.10206561360874848, | |
| "grad_norm": 2.716110944747925, | |
| "learning_rate": 0.00029495596607919305, | |
| "loss": 0.2997, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.10530579181855002, | |
| "eval_loss": 0.3230917453765869, | |
| "eval_runtime": 88.9179, | |
| "eval_samples_per_second": 5.623, | |
| "eval_steps_per_second": 0.709, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.11340623734305387, | |
| "grad_norm": 2.5921218395233154, | |
| "learning_rate": 0.00029345584767191685, | |
| "loss": 0.3052, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.12474686107735926, | |
| "grad_norm": 2.5036072731018066, | |
| "learning_rate": 0.0002917655810607161, | |
| "loss": 0.3196, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.12636695018226002, | |
| "eval_loss": 0.3350300192832947, | |
| "eval_runtime": 89.9537, | |
| "eval_samples_per_second": 5.558, | |
| "eval_steps_per_second": 0.7, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.13608748481166463, | |
| "grad_norm": 2.4138002395629883, | |
| "learning_rate": 0.0002898874066642667, | |
| "loss": 0.3225, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.14742810854597002, | |
| "grad_norm": 2.376221179962158, | |
| "learning_rate": 0.00028782381396971003, | |
| "loss": 0.3263, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.14742810854597002, | |
| "eval_loss": 0.3406391739845276, | |
| "eval_runtime": 12.3974, | |
| "eval_samples_per_second": 40.331, | |
| "eval_steps_per_second": 5.082, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.1587687322802754, | |
| "grad_norm": 2.2181365489959717, | |
| "learning_rate": 0.00028557753823288173, | |
| "loss": 0.3148, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 0.16848926690968002, | |
| "eval_loss": 0.3401181101799011, | |
| "eval_runtime": 90.3528, | |
| "eval_samples_per_second": 5.534, | |
| "eval_steps_per_second": 0.697, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.1701093560145808, | |
| "grad_norm": 2.313964366912842, | |
| "learning_rate": 0.0002831515568527781, | |
| "loss": 0.3245, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.1814499797488862, | |
| "grad_norm": 2.2569937705993652, | |
| "learning_rate": 0.00028054908542506627, | |
| "loss": 0.3297, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.18955042527339003, | |
| "eval_loss": 0.3456376791000366, | |
| "eval_runtime": 88.1843, | |
| "eval_samples_per_second": 5.67, | |
| "eval_steps_per_second": 0.714, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 0.19279060348319157, | |
| "grad_norm": 2.462419033050537, | |
| "learning_rate": 0.00027777357347986823, | |
| "loss": 0.3189, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 0.20413122721749696, | |
| "grad_norm": 2.143909215927124, | |
| "learning_rate": 0.00027482869990946986, | |
| "loss": 0.3221, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 0.21061158363710003, | |
| "eval_loss": 0.3476516604423523, | |
| "eval_runtime": 12.3771, | |
| "eval_samples_per_second": 40.397, | |
| "eval_steps_per_second": 5.09, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.21547185095180235, | |
| "grad_norm": 1.9863917827606201, | |
| "learning_rate": 0.0002717183680920135, | |
| "loss": 0.3263, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 0.22681247468610774, | |
| "grad_norm": 2.3500940799713135, | |
| "learning_rate": 0.00026844670071763906, | |
| "loss": 0.3359, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.23167274200081003, | |
| "eval_loss": 0.3490840792655945, | |
| "eval_runtime": 92.3634, | |
| "eval_samples_per_second": 5.413, | |
| "eval_steps_per_second": 0.682, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 0.23815309842041313, | |
| "grad_norm": 2.0207138061523438, | |
| "learning_rate": 0.00026501803432393037, | |
| "loss": 0.327, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 0.24949372215471852, | |
| "grad_norm": 2.1643483638763428, | |
| "learning_rate": 0.00026143691354791145, | |
| "loss": 0.3296, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 0.25273390036452004, | |
| "eval_loss": 0.3398853838443756, | |
| "eval_runtime": 91.6137, | |
| "eval_samples_per_second": 5.458, | |
| "eval_steps_per_second": 0.688, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.2608343458890239, | |
| "grad_norm": 2.1574275493621826, | |
| "learning_rate": 0.00025770808510220956, | |
| "loss": 0.3279, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 0.27217496962332927, | |
| "grad_norm": 2.1672723293304443, | |
| "learning_rate": 0.00025383649148337105, | |
| "loss": 0.3361, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.27379505872823007, | |
| "eval_loss": 0.34163641929626465, | |
| "eval_runtime": 12.4303, | |
| "eval_samples_per_second": 40.224, | |
| "eval_steps_per_second": 5.068, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 0.28351559335763465, | |
| "grad_norm": 2.2144994735717773, | |
| "learning_rate": 0.0002498272644206695, | |
| "loss": 0.3341, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.29485621709194004, | |
| "grad_norm": 2.0570850372314453, | |
| "learning_rate": 0.0002456857180740884, | |
| "loss": 0.3187, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 0.29485621709194004, | |
| "eval_loss": 0.3376466631889343, | |
| "eval_runtime": 74.564, | |
| "eval_samples_per_second": 6.706, | |
| "eval_steps_per_second": 0.845, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 0.30619684082624543, | |
| "grad_norm": 2.23764967918396, | |
| "learning_rate": 0.0002414173419904956, | |
| "loss": 0.3285, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 0.3159173754556501, | |
| "eval_loss": 0.33701062202453613, | |
| "eval_runtime": 85.9424, | |
| "eval_samples_per_second": 5.818, | |
| "eval_steps_per_second": 0.733, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.3175374645605508, | |
| "grad_norm": 1.9056419134140015, | |
| "learning_rate": 0.00023702779382734566, | |
| "loss": 0.3197, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.3288780882948562, | |
| "grad_norm": 2.03104305267334, | |
| "learning_rate": 0.0002325228918535541, | |
| "loss": 0.3189, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 0.33697853381936005, | |
| "eval_loss": 0.33059319853782654, | |
| "eval_runtime": 12.4539, | |
| "eval_samples_per_second": 40.148, | |
| "eval_steps_per_second": 5.059, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.3402187120291616, | |
| "grad_norm": 2.170870780944824, | |
| "learning_rate": 0.00022790860723748442, | |
| "loss": 0.3232, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.351559335763467, | |
| "grad_norm": 1.942113995552063, | |
| "learning_rate": 0.00022319105613226921, | |
| "loss": 0.3154, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 0.3580396921830701, | |
| "eval_loss": 0.32931268215179443, | |
| "eval_runtime": 86.8422, | |
| "eval_samples_per_second": 5.758, | |
| "eval_steps_per_second": 0.725, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 0.3628999594977724, | |
| "grad_norm": 1.8157175779342651, | |
| "learning_rate": 0.00021837649156895706, | |
| "loss": 0.3205, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.37424058323207776, | |
| "grad_norm": 2.035099983215332, | |
| "learning_rate": 0.00021347129516822945, | |
| "loss": 0.3149, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 0.37910085054678005, | |
| "eval_loss": 0.3263063430786133, | |
| "eval_runtime": 78.1253, | |
| "eval_samples_per_second": 6.4, | |
| "eval_steps_per_second": 0.806, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 0.38558120696638315, | |
| "grad_norm": 2.07989764213562, | |
| "learning_rate": 0.00020848196868167505, | |
| "loss": 0.3115, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 0.39692183070068854, | |
| "grad_norm": 1.8494740724563599, | |
| "learning_rate": 0.000203415125373832, | |
| "loss": 0.3099, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.4001620089104901, | |
| "eval_loss": 0.32076019048690796, | |
| "eval_runtime": 12.4446, | |
| "eval_samples_per_second": 40.178, | |
| "eval_steps_per_second": 5.062, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 0.4082624544349939, | |
| "grad_norm": 2.0092668533325195, | |
| "learning_rate": 0.00019827748125642242, | |
| "loss": 0.3131, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.4196030781692993, | |
| "grad_norm": 2.0937247276306152, | |
| "learning_rate": 0.0001930758461863965, | |
| "loss": 0.3089, | |
| "step": 259 | |
| }, | |
| { | |
| "epoch": 0.42122316727420006, | |
| "eval_loss": 0.3143251836299896, | |
| "eval_runtime": 77.5222, | |
| "eval_samples_per_second": 6.45, | |
| "eval_steps_per_second": 0.813, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.4309437019036047, | |
| "grad_norm": 2.0988640785217285, | |
| "learning_rate": 0.0001878171148395872, | |
| "loss": 0.3124, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 0.4422843256379101, | |
| "grad_norm": 2.4692976474761963, | |
| "learning_rate": 0.00018250825757193848, | |
| "loss": 0.3125, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 0.4422843256379101, | |
| "eval_loss": 0.31036385893821716, | |
| "eval_runtime": 84.8095, | |
| "eval_samples_per_second": 5.896, | |
| "eval_steps_per_second": 0.743, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 0.4536249493722155, | |
| "grad_norm": 1.8889740705490112, | |
| "learning_rate": 0.0001771563111804211, | |
| "loss": 0.2959, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.46334548400162007, | |
| "eval_loss": 0.30606991052627563, | |
| "eval_runtime": 12.4497, | |
| "eval_samples_per_second": 40.162, | |
| "eval_steps_per_second": 5.06, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 0.46496557310652087, | |
| "grad_norm": 2.023824691772461, | |
| "learning_rate": 0.0001717683695758819, | |
| "loss": 0.299, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 0.47630619684082626, | |
| "grad_norm": 2.0295095443725586, | |
| "learning_rate": 0.00016635157438018983, | |
| "loss": 0.3042, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 0.4844066423653301, | |
| "eval_loss": 0.2992589473724365, | |
| "eval_runtime": 75.8225, | |
| "eval_samples_per_second": 6.594, | |
| "eval_steps_per_second": 0.831, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 0.48764682057513165, | |
| "grad_norm": 1.946453332901001, | |
| "learning_rate": 0.0001609131054601416, | |
| "loss": 0.294, | |
| "step": 301 | |
| }, | |
| { | |
| "epoch": 0.49898744430943703, | |
| "grad_norm": 1.9718040227890015, | |
| "learning_rate": 0.00015546017141067432, | |
| "loss": 0.2829, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.5054678007290401, | |
| "eval_loss": 0.294008731842041, | |
| "eval_runtime": 82.5163, | |
| "eval_samples_per_second": 6.059, | |
| "eval_steps_per_second": 0.763, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.5103280680437424, | |
| "grad_norm": 1.8706125020980835, | |
| "learning_rate": 0.00015, | |
| "loss": 0.2802, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.5216686917780478, | |
| "grad_norm": 1.961658000946045, | |
| "learning_rate": 0.0001445398285893257, | |
| "loss": 0.2832, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 0.5265289590927501, | |
| "eval_loss": 0.2878178656101227, | |
| "eval_runtime": 12.4298, | |
| "eval_samples_per_second": 40.226, | |
| "eval_steps_per_second": 5.068, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.5330093155123532, | |
| "grad_norm": 1.774943232536316, | |
| "learning_rate": 0.0001390868945398584, | |
| "loss": 0.2763, | |
| "step": 329 | |
| }, | |
| { | |
| "epoch": 0.5443499392466585, | |
| "grad_norm": 2.2113044261932373, | |
| "learning_rate": 0.00013364842561981014, | |
| "loss": 0.2715, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.5475901174564601, | |
| "eval_loss": 0.2820768654346466, | |
| "eval_runtime": 85.9045, | |
| "eval_samples_per_second": 5.82, | |
| "eval_steps_per_second": 0.733, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 0.555690562980964, | |
| "grad_norm": 1.9216161966323853, | |
| "learning_rate": 0.00012823163042411807, | |
| "loss": 0.2706, | |
| "step": 343 | |
| }, | |
| { | |
| "epoch": 0.5670311867152693, | |
| "grad_norm": 2.2038073539733887, | |
| "learning_rate": 0.0001228436888195789, | |
| "loss": 0.2702, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.56865127582017, | |
| "eval_loss": 0.27528542280197144, | |
| "eval_runtime": 76.8796, | |
| "eval_samples_per_second": 6.504, | |
| "eval_steps_per_second": 0.819, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 0.5783718104495748, | |
| "grad_norm": 2.00290846824646, | |
| "learning_rate": 0.00011749174242806152, | |
| "loss": 0.2704, | |
| "step": 357 | |
| }, | |
| { | |
| "epoch": 0.5897124341838801, | |
| "grad_norm": 1.8399847745895386, | |
| "learning_rate": 0.00011218288516041279, | |
| "loss": 0.2687, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.5897124341838801, | |
| "eval_loss": 0.2687266170978546, | |
| "eval_runtime": 12.4501, | |
| "eval_samples_per_second": 40.16, | |
| "eval_steps_per_second": 5.06, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.6010530579181855, | |
| "grad_norm": 1.754230260848999, | |
| "learning_rate": 0.00010692415381360349, | |
| "loss": 0.2604, | |
| "step": 371 | |
| }, | |
| { | |
| "epoch": 0.6107735925475901, | |
| "eval_loss": 0.26285678148269653, | |
| "eval_runtime": 89.7111, | |
| "eval_samples_per_second": 5.573, | |
| "eval_steps_per_second": 0.702, | |
| "step": 377 | |
| }, | |
| { | |
| "epoch": 0.6123936816524909, | |
| "grad_norm": 2.0383028984069824, | |
| "learning_rate": 0.00010172251874357757, | |
| "loss": 0.2565, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 0.6237343053867963, | |
| "grad_norm": 1.9088571071624756, | |
| "learning_rate": 9.658487462616794e-05, | |
| "loss": 0.252, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.6318347509113001, | |
| "eval_loss": 0.25791773200035095, | |
| "eval_runtime": 88.6204, | |
| "eval_samples_per_second": 5.642, | |
| "eval_steps_per_second": 0.711, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.6350749291211016, | |
| "grad_norm": 1.8957221508026123, | |
| "learning_rate": 9.151803131832493e-05, | |
| "loss": 0.2709, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 0.6464155528554071, | |
| "grad_norm": 1.6993653774261475, | |
| "learning_rate": 8.652870483177049e-05, | |
| "loss": 0.2537, | |
| "step": 399 | |
| }, | |
| { | |
| "epoch": 0.6528959092750102, | |
| "eval_loss": 0.2528708577156067, | |
| "eval_runtime": 12.4911, | |
| "eval_samples_per_second": 40.028, | |
| "eval_steps_per_second": 5.044, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 0.6577561765897124, | |
| "grad_norm": 1.844677209854126, | |
| "learning_rate": 8.162350843104291e-05, | |
| "loss": 0.2468, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 0.6690968003240179, | |
| "grad_norm": 1.943462610244751, | |
| "learning_rate": 7.680894386773072e-05, | |
| "loss": 0.2535, | |
| "step": 413 | |
| }, | |
| { | |
| "epoch": 0.6739570676387201, | |
| "eval_loss": 0.24768707156181335, | |
| "eval_runtime": 89.239, | |
| "eval_samples_per_second": 5.603, | |
| "eval_steps_per_second": 0.706, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 0.6804374240583232, | |
| "grad_norm": 1.9346702098846436, | |
| "learning_rate": 7.209139276251558e-05, | |
| "loss": 0.2472, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.6917780477926286, | |
| "grad_norm": 1.684934377670288, | |
| "learning_rate": 6.747710814644589e-05, | |
| "loss": 0.2442, | |
| "step": 427 | |
| }, | |
| { | |
| "epoch": 0.6950182260024301, | |
| "eval_loss": 0.24250428378582, | |
| "eval_runtime": 90.1456, | |
| "eval_samples_per_second": 5.547, | |
| "eval_steps_per_second": 0.699, | |
| "step": 429 | |
| }, | |
| { | |
| "epoch": 0.703118671526934, | |
| "grad_norm": 1.652444839477539, | |
| "learning_rate": 6.297220617265435e-05, | |
| "loss": 0.233, | |
| "step": 434 | |
| }, | |
| { | |
| "epoch": 0.7144592952612394, | |
| "grad_norm": 1.749850869178772, | |
| "learning_rate": 5.858265800950438e-05, | |
| "loss": 0.2451, | |
| "step": 441 | |
| }, | |
| { | |
| "epoch": 0.7160793843661402, | |
| "eval_loss": 0.237756609916687, | |
| "eval_runtime": 12.447, | |
| "eval_samples_per_second": 40.17, | |
| "eval_steps_per_second": 5.061, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 0.7257999189955447, | |
| "grad_norm": 1.607130527496338, | |
| "learning_rate": 5.4314281925911634e-05, | |
| "loss": 0.2291, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 0.7371405427298502, | |
| "grad_norm": 1.7240676879882812, | |
| "learning_rate": 5.0172735579330526e-05, | |
| "loss": 0.2275, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.7371405427298502, | |
| "eval_loss": 0.23375177383422852, | |
| "eval_runtime": 88.4413, | |
| "eval_samples_per_second": 5.653, | |
| "eval_steps_per_second": 0.712, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.7484811664641555, | |
| "grad_norm": 1.821494698524475, | |
| "learning_rate": 4.616350851662895e-05, | |
| "loss": 0.2288, | |
| "step": 462 | |
| }, | |
| { | |
| "epoch": 0.7582017010935601, | |
| "eval_loss": 0.23097103834152222, | |
| "eval_runtime": 91.1616, | |
| "eval_samples_per_second": 5.485, | |
| "eval_steps_per_second": 0.691, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 0.759821790198461, | |
| "grad_norm": 1.6781857013702393, | |
| "learning_rate": 4.229191489779047e-05, | |
| "loss": 0.2291, | |
| "step": 469 | |
| }, | |
| { | |
| "epoch": 0.7711624139327663, | |
| "grad_norm": 1.8011544942855835, | |
| "learning_rate": 3.8563086452088506e-05, | |
| "loss": 0.2323, | |
| "step": 476 | |
| }, | |
| { | |
| "epoch": 0.7792628594572701, | |
| "eval_loss": 0.22944015264511108, | |
| "eval_runtime": 12.4573, | |
| "eval_samples_per_second": 40.137, | |
| "eval_steps_per_second": 5.057, | |
| "step": 481 | |
| }, | |
| { | |
| "epoch": 0.7825030376670717, | |
| "grad_norm": 1.6624199151992798, | |
| "learning_rate": 3.498196567606959e-05, | |
| "loss": 0.2238, | |
| "step": 483 | |
| }, | |
| { | |
| "epoch": 0.7938436614013771, | |
| "grad_norm": 1.7592713832855225, | |
| "learning_rate": 3.1553299282360966e-05, | |
| "loss": 0.2254, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.8003240178209802, | |
| "eval_loss": 0.225956991314888, | |
| "eval_runtime": 79.2699, | |
| "eval_samples_per_second": 6.308, | |
| "eval_steps_per_second": 0.795, | |
| "step": 494 | |
| }, | |
| { | |
| "epoch": 0.8051842851356824, | |
| "grad_norm": 1.757391333580017, | |
| "learning_rate": 2.828163190798644e-05, | |
| "loss": 0.2203, | |
| "step": 497 | |
| }, | |
| { | |
| "epoch": 0.8165249088699879, | |
| "grad_norm": 1.7900511026382446, | |
| "learning_rate": 2.5171300090530106e-05, | |
| "loss": 0.2142, | |
| "step": 504 | |
| }, | |
| { | |
| "epoch": 0.8213851761846902, | |
| "eval_loss": 0.22206208109855652, | |
| "eval_runtime": 85.0506, | |
| "eval_samples_per_second": 5.879, | |
| "eval_steps_per_second": 0.741, | |
| "step": 507 | |
| }, | |
| { | |
| "epoch": 0.8278655326042932, | |
| "grad_norm": 1.7866923809051514, | |
| "learning_rate": 2.2226426520131734e-05, | |
| "loss": 0.2147, | |
| "step": 511 | |
| }, | |
| { | |
| "epoch": 0.8392061563385986, | |
| "grad_norm": 1.5760256052017212, | |
| "learning_rate": 1.9450914574933725e-05, | |
| "loss": 0.219, | |
| "step": 518 | |
| }, | |
| { | |
| "epoch": 0.8424463345484001, | |
| "eval_loss": 0.21945638954639435, | |
| "eval_runtime": 12.4577, | |
| "eval_samples_per_second": 40.136, | |
| "eval_steps_per_second": 5.057, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.850546780072904, | |
| "grad_norm": 1.7267919778823853, | |
| "learning_rate": 1.6848443147221828e-05, | |
| "loss": 0.2101, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.8618874038072094, | |
| "grad_norm": 1.6345723867416382, | |
| "learning_rate": 1.4422461767118233e-05, | |
| "loss": 0.2133, | |
| "step": 532 | |
| }, | |
| { | |
| "epoch": 0.8635074929121102, | |
| "eval_loss": 0.2179647982120514, | |
| "eval_runtime": 85.1994, | |
| "eval_samples_per_second": 5.869, | |
| "eval_steps_per_second": 0.739, | |
| "step": 533 | |
| }, | |
| { | |
| "epoch": 0.8732280275415147, | |
| "grad_norm": 1.4804401397705078, | |
| "learning_rate": 1.2176186030289936e-05, | |
| "loss": 0.2136, | |
| "step": 539 | |
| }, | |
| { | |
| "epoch": 0.8845686512758202, | |
| "grad_norm": 1.593434453010559, | |
| "learning_rate": 1.011259333573326e-05, | |
| "loss": 0.2095, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 0.8845686512758202, | |
| "eval_loss": 0.21640805900096893, | |
| "eval_runtime": 75.9951, | |
| "eval_samples_per_second": 6.579, | |
| "eval_steps_per_second": 0.829, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 0.8959092750101255, | |
| "grad_norm": 1.5899996757507324, | |
| "learning_rate": 8.234418939283866e-06, | |
| "loss": 0.2067, | |
| "step": 553 | |
| }, | |
| { | |
| "epoch": 0.9056298096395302, | |
| "eval_loss": 0.2154710441827774, | |
| "eval_runtime": 12.505, | |
| "eval_samples_per_second": 39.984, | |
| "eval_steps_per_second": 5.038, | |
| "step": 559 | |
| }, | |
| { | |
| "epoch": 0.907249898744431, | |
| "grad_norm": 1.567068099975586, | |
| "learning_rate": 6.544152328083152e-06, | |
| "loss": 0.2117, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.9185905224787363, | |
| "grad_norm": 1.676525354385376, | |
| "learning_rate": 5.044033920806933e-06, | |
| "loss": 0.2073, | |
| "step": 567 | |
| }, | |
| { | |
| "epoch": 0.9266909680032401, | |
| "eval_loss": 0.21462255716323853, | |
| "eval_runtime": 87.8678, | |
| "eval_samples_per_second": 5.69, | |
| "eval_steps_per_second": 0.717, | |
| "step": 572 | |
| }, | |
| { | |
| "epoch": 0.9299311462130417, | |
| "grad_norm": 1.5873310565948486, | |
| "learning_rate": 3.7360520980297514e-06, | |
| "loss": 0.208, | |
| "step": 574 | |
| }, | |
| { | |
| "epoch": 0.9412717699473471, | |
| "grad_norm": 1.7347217798233032, | |
| "learning_rate": 2.6219405666614402e-06, | |
| "loss": 0.2124, | |
| "step": 581 | |
| }, | |
| { | |
| "epoch": 0.9477521263669502, | |
| "eval_loss": 0.21402177214622498, | |
| "eval_runtime": 87.5614, | |
| "eval_samples_per_second": 5.71, | |
| "eval_steps_per_second": 0.719, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 0.9526123936816525, | |
| "grad_norm": 1.8601125478744507, | |
| "learning_rate": 1.7031760619491353e-06, | |
| "loss": 0.2076, | |
| "step": 588 | |
| }, | |
| { | |
| "epoch": 0.9639530174159578, | |
| "grad_norm": 1.6909899711608887, | |
| "learning_rate": 9.809763900905875e-07, | |
| "loss": 0.2115, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 0.9688132847306602, | |
| "eval_loss": 0.21379441022872925, | |
| "eval_runtime": 12.4885, | |
| "eval_samples_per_second": 40.037, | |
| "eval_steps_per_second": 5.045, | |
| "step": 598 | |
| }, | |
| { | |
| "epoch": 0.9752936411502633, | |
| "grad_norm": 1.5291049480438232, | |
| "learning_rate": 4.562988140535073e-07, | |
| "loss": 0.2071, | |
| "step": 602 | |
| }, | |
| { | |
| "epoch": 0.9866342648845686, | |
| "grad_norm": 1.775606632232666, | |
| "learning_rate": 1.298387847403437e-07, | |
| "loss": 0.2127, | |
| "step": 609 | |
| }, | |
| { | |
| "epoch": 0.9898744430943702, | |
| "eval_loss": 0.21362636983394623, | |
| "eval_runtime": 90.3482, | |
| "eval_samples_per_second": 5.534, | |
| "eval_steps_per_second": 0.697, | |
| "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": 7.406584397518602e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |