|
{ |
|
"best_metric": 0.18765884637832642, |
|
"best_model_checkpoint": "/home/datta0/models/lora_final/Qwen2-7B_metamath_default/checkpoint-13", |
|
"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": 4.52362585067749, |
|
"learning_rate": 2.3076923076923076e-05, |
|
"loss": 0.2546, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.011340623734305387, |
|
"grad_norm": 1.5377665758132935, |
|
"learning_rate": 0.00016153846153846153, |
|
"loss": 0.1753, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 0.021061158363710003, |
|
"eval_loss": 0.18765884637832642, |
|
"eval_runtime": 12.39, |
|
"eval_samples_per_second": 40.355, |
|
"eval_steps_per_second": 5.085, |
|
"step": 13 |
|
}, |
|
{ |
|
"epoch": 0.022681247468610773, |
|
"grad_norm": 2.192612409591675, |
|
"learning_rate": 0.0002999979709808197, |
|
"loss": 0.1716, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 0.03402187120291616, |
|
"grad_norm": 2.496940851211548, |
|
"learning_rate": 0.0002998701612152596, |
|
"loss": 0.2034, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 0.042122316727420006, |
|
"eval_loss": 0.2530185580253601, |
|
"eval_runtime": 81.8512, |
|
"eval_samples_per_second": 6.109, |
|
"eval_steps_per_second": 0.77, |
|
"step": 26 |
|
}, |
|
{ |
|
"epoch": 0.04536249493722155, |
|
"grad_norm": 2.3953869342803955, |
|
"learning_rate": 0.0002995437011859465, |
|
"loss": 0.2398, |
|
"step": 28 |
|
}, |
|
{ |
|
"epoch": 0.056703118671526935, |
|
"grad_norm": 2.8134021759033203, |
|
"learning_rate": 0.00029901902360990936, |
|
"loss": 0.261, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 0.06318347509113001, |
|
"eval_loss": 0.2844119071960449, |
|
"eval_runtime": 86.8656, |
|
"eval_samples_per_second": 5.756, |
|
"eval_steps_per_second": 0.725, |
|
"step": 39 |
|
}, |
|
{ |
|
"epoch": 0.06804374240583232, |
|
"grad_norm": 2.448476791381836, |
|
"learning_rate": 0.00029829682393805085, |
|
"loss": 0.2623, |
|
"step": 42 |
|
}, |
|
{ |
|
"epoch": 0.0793843661401377, |
|
"grad_norm": 2.409214973449707, |
|
"learning_rate": 0.0002973780594333385, |
|
"loss": 0.2864, |
|
"step": 49 |
|
}, |
|
{ |
|
"epoch": 0.08424463345484001, |
|
"eval_loss": 0.3106594681739807, |
|
"eval_runtime": 12.4045, |
|
"eval_samples_per_second": 40.308, |
|
"eval_steps_per_second": 5.079, |
|
"step": 52 |
|
}, |
|
{ |
|
"epoch": 0.0907249898744431, |
|
"grad_norm": 2.424531936645508, |
|
"learning_rate": 0.00029626394790197025, |
|
"loss": 0.3005, |
|
"step": 56 |
|
}, |
|
{ |
|
"epoch": 0.10206561360874848, |
|
"grad_norm": 2.680947780609131, |
|
"learning_rate": 0.00029495596607919305, |
|
"loss": 0.2984, |
|
"step": 63 |
|
}, |
|
{ |
|
"epoch": 0.10530579181855002, |
|
"eval_loss": 0.32644107937812805, |
|
"eval_runtime": 85.363, |
|
"eval_samples_per_second": 5.857, |
|
"eval_steps_per_second": 0.738, |
|
"step": 65 |
|
}, |
|
{ |
|
"epoch": 0.11340623734305387, |
|
"grad_norm": 2.4300179481506348, |
|
"learning_rate": 0.00029345584767191685, |
|
"loss": 0.306, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.12474686107735926, |
|
"grad_norm": 2.570291519165039, |
|
"learning_rate": 0.0002917655810607161, |
|
"loss": 0.318, |
|
"step": 77 |
|
}, |
|
{ |
|
"epoch": 0.12636695018226002, |
|
"eval_loss": 0.33606359362602234, |
|
"eval_runtime": 75.386, |
|
"eval_samples_per_second": 6.633, |
|
"eval_steps_per_second": 0.836, |
|
"step": 78 |
|
}, |
|
{ |
|
"epoch": 0.13608748481166463, |
|
"grad_norm": 2.4007227420806885, |
|
"learning_rate": 0.0002898874066642667, |
|
"loss": 0.3192, |
|
"step": 84 |
|
}, |
|
{ |
|
"epoch": 0.14742810854597002, |
|
"grad_norm": 2.463517427444458, |
|
"learning_rate": 0.00028782381396971003, |
|
"loss": 0.3257, |
|
"step": 91 |
|
}, |
|
{ |
|
"epoch": 0.14742810854597002, |
|
"eval_loss": 0.3400190472602844, |
|
"eval_runtime": 12.4934, |
|
"eval_samples_per_second": 40.021, |
|
"eval_steps_per_second": 5.043, |
|
"step": 91 |
|
}, |
|
{ |
|
"epoch": 0.1587687322802754, |
|
"grad_norm": 2.301016330718994, |
|
"learning_rate": 0.00028557753823288173, |
|
"loss": 0.3139, |
|
"step": 98 |
|
}, |
|
{ |
|
"epoch": 0.16848926690968002, |
|
"eval_loss": 0.33769935369491577, |
|
"eval_runtime": 90.0772, |
|
"eval_samples_per_second": 5.551, |
|
"eval_steps_per_second": 0.699, |
|
"step": 104 |
|
}, |
|
{ |
|
"epoch": 0.1701093560145808, |
|
"grad_norm": 2.232789993286133, |
|
"learning_rate": 0.0002831515568527781, |
|
"loss": 0.326, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 0.1814499797488862, |
|
"grad_norm": 2.174959897994995, |
|
"learning_rate": 0.00028054908542506627, |
|
"loss": 0.327, |
|
"step": 112 |
|
}, |
|
{ |
|
"epoch": 0.18955042527339003, |
|
"eval_loss": 0.3428936302661896, |
|
"eval_runtime": 89.6799, |
|
"eval_samples_per_second": 5.575, |
|
"eval_steps_per_second": 0.702, |
|
"step": 117 |
|
}, |
|
{ |
|
"epoch": 0.19279060348319157, |
|
"grad_norm": 2.3661577701568604, |
|
"learning_rate": 0.00027777357347986823, |
|
"loss": 0.319, |
|
"step": 119 |
|
}, |
|
{ |
|
"epoch": 0.20413122721749696, |
|
"grad_norm": 2.149407386779785, |
|
"learning_rate": 0.00027482869990946986, |
|
"loss": 0.3219, |
|
"step": 126 |
|
}, |
|
{ |
|
"epoch": 0.21061158363710003, |
|
"eval_loss": 0.3457578420639038, |
|
"eval_runtime": 12.3968, |
|
"eval_samples_per_second": 40.333, |
|
"eval_steps_per_second": 5.082, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.21547185095180235, |
|
"grad_norm": 2.0185813903808594, |
|
"learning_rate": 0.0002717183680920135, |
|
"loss": 0.3251, |
|
"step": 133 |
|
}, |
|
{ |
|
"epoch": 0.22681247468610774, |
|
"grad_norm": 2.2934110164642334, |
|
"learning_rate": 0.00026844670071763906, |
|
"loss": 0.3325, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.23167274200081003, |
|
"eval_loss": 0.34575140476226807, |
|
"eval_runtime": 90.2436, |
|
"eval_samples_per_second": 5.541, |
|
"eval_steps_per_second": 0.698, |
|
"step": 143 |
|
}, |
|
{ |
|
"epoch": 0.23815309842041313, |
|
"grad_norm": 2.0038561820983887, |
|
"learning_rate": 0.00026501803432393037, |
|
"loss": 0.3276, |
|
"step": 147 |
|
}, |
|
{ |
|
"epoch": 0.24949372215471852, |
|
"grad_norm": 2.029125690460205, |
|
"learning_rate": 0.00026143691354791145, |
|
"loss": 0.3273, |
|
"step": 154 |
|
}, |
|
{ |
|
"epoch": 0.25273390036452004, |
|
"eval_loss": 0.34069105982780457, |
|
"eval_runtime": 82.809, |
|
"eval_samples_per_second": 6.038, |
|
"eval_steps_per_second": 0.761, |
|
"step": 156 |
|
}, |
|
{ |
|
"epoch": 0.2608343458890239, |
|
"grad_norm": 2.2533740997314453, |
|
"learning_rate": 0.00025770808510220956, |
|
"loss": 0.3313, |
|
"step": 161 |
|
}, |
|
{ |
|
"epoch": 0.27217496962332927, |
|
"grad_norm": 2.204942226409912, |
|
"learning_rate": 0.00025383649148337105, |
|
"loss": 0.3328, |
|
"step": 168 |
|
}, |
|
{ |
|
"epoch": 0.27379505872823007, |
|
"eval_loss": 0.3394912779331207, |
|
"eval_runtime": 12.4532, |
|
"eval_samples_per_second": 40.15, |
|
"eval_steps_per_second": 5.059, |
|
"step": 169 |
|
}, |
|
{ |
|
"epoch": 0.28351559335763465, |
|
"grad_norm": 2.162797689437866, |
|
"learning_rate": 0.0002498272644206695, |
|
"loss": 0.3332, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 0.29485621709194004, |
|
"grad_norm": 2.1584229469299316, |
|
"learning_rate": 0.0002456857180740884, |
|
"loss": 0.32, |
|
"step": 182 |
|
}, |
|
{ |
|
"epoch": 0.29485621709194004, |
|
"eval_loss": 0.33717018365859985, |
|
"eval_runtime": 77.5777, |
|
"eval_samples_per_second": 6.445, |
|
"eval_steps_per_second": 0.812, |
|
"step": 182 |
|
}, |
|
{ |
|
"epoch": 0.30619684082624543, |
|
"grad_norm": 2.1494665145874023, |
|
"learning_rate": 0.0002414173419904956, |
|
"loss": 0.3293, |
|
"step": 189 |
|
}, |
|
{ |
|
"epoch": 0.3159173754556501, |
|
"eval_loss": 0.33609288930892944, |
|
"eval_runtime": 93.3699, |
|
"eval_samples_per_second": 5.355, |
|
"eval_steps_per_second": 0.675, |
|
"step": 195 |
|
}, |
|
{ |
|
"epoch": 0.3175374645605508, |
|
"grad_norm": 1.9434106349945068, |
|
"learning_rate": 0.00023702779382734566, |
|
"loss": 0.32, |
|
"step": 196 |
|
}, |
|
{ |
|
"epoch": 0.3288780882948562, |
|
"grad_norm": 1.9602625370025635, |
|
"learning_rate": 0.0002325228918535541, |
|
"loss": 0.3182, |
|
"step": 203 |
|
}, |
|
{ |
|
"epoch": 0.33697853381936005, |
|
"eval_loss": 0.3306241035461426, |
|
"eval_runtime": 12.4056, |
|
"eval_samples_per_second": 40.304, |
|
"eval_steps_per_second": 5.078, |
|
"step": 208 |
|
}, |
|
{ |
|
"epoch": 0.3402187120291616, |
|
"grad_norm": 2.2286486625671387, |
|
"learning_rate": 0.00022790860723748442, |
|
"loss": 0.3218, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.351559335763467, |
|
"grad_norm": 2.0589637756347656, |
|
"learning_rate": 0.00022319105613226921, |
|
"loss": 0.3148, |
|
"step": 217 |
|
}, |
|
{ |
|
"epoch": 0.3580396921830701, |
|
"eval_loss": 0.3287665545940399, |
|
"eval_runtime": 77.8166, |
|
"eval_samples_per_second": 6.425, |
|
"eval_steps_per_second": 0.81, |
|
"step": 221 |
|
}, |
|
{ |
|
"epoch": 0.3628999594977724, |
|
"grad_norm": 1.8553071022033691, |
|
"learning_rate": 0.00021837649156895706, |
|
"loss": 0.3206, |
|
"step": 224 |
|
}, |
|
{ |
|
"epoch": 0.37424058323207776, |
|
"grad_norm": 2.019782304763794, |
|
"learning_rate": 0.00021347129516822945, |
|
"loss": 0.3128, |
|
"step": 231 |
|
}, |
|
{ |
|
"epoch": 0.37910085054678005, |
|
"eval_loss": 0.32550591230392456, |
|
"eval_runtime": 78.3487, |
|
"eval_samples_per_second": 6.382, |
|
"eval_steps_per_second": 0.804, |
|
"step": 234 |
|
}, |
|
{ |
|
"epoch": 0.38558120696638315, |
|
"grad_norm": 2.100358247756958, |
|
"learning_rate": 0.00020848196868167505, |
|
"loss": 0.3111, |
|
"step": 238 |
|
}, |
|
{ |
|
"epoch": 0.39692183070068854, |
|
"grad_norm": 1.8522151708602905, |
|
"learning_rate": 0.000203415125373832, |
|
"loss": 0.3086, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 0.4001620089104901, |
|
"eval_loss": 0.3193366825580597, |
|
"eval_runtime": 12.4578, |
|
"eval_samples_per_second": 40.136, |
|
"eval_steps_per_second": 5.057, |
|
"step": 247 |
|
}, |
|
{ |
|
"epoch": 0.4082624544349939, |
|
"grad_norm": 2.0124332904815674, |
|
"learning_rate": 0.00019827748125642242, |
|
"loss": 0.3119, |
|
"step": 252 |
|
}, |
|
{ |
|
"epoch": 0.4196030781692993, |
|
"grad_norm": 2.0748066902160645, |
|
"learning_rate": 0.0001930758461863965, |
|
"loss": 0.306, |
|
"step": 259 |
|
}, |
|
{ |
|
"epoch": 0.42122316727420006, |
|
"eval_loss": 0.31428423523902893, |
|
"eval_runtime": 92.6512, |
|
"eval_samples_per_second": 5.397, |
|
"eval_steps_per_second": 0.68, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.4309437019036047, |
|
"grad_norm": 2.1250007152557373, |
|
"learning_rate": 0.0001878171148395872, |
|
"loss": 0.3068, |
|
"step": 266 |
|
}, |
|
{ |
|
"epoch": 0.4422843256379101, |
|
"grad_norm": 2.4317753314971924, |
|
"learning_rate": 0.00018250825757193848, |
|
"loss": 0.3122, |
|
"step": 273 |
|
}, |
|
{ |
|
"epoch": 0.4422843256379101, |
|
"eval_loss": 0.31197118759155273, |
|
"eval_runtime": 82.8282, |
|
"eval_samples_per_second": 6.037, |
|
"eval_steps_per_second": 0.761, |
|
"step": 273 |
|
}, |
|
{ |
|
"epoch": 0.4536249493722155, |
|
"grad_norm": 1.870219349861145, |
|
"learning_rate": 0.0001771563111804211, |
|
"loss": 0.2982, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.46334548400162007, |
|
"eval_loss": 0.30591174960136414, |
|
"eval_runtime": 12.4327, |
|
"eval_samples_per_second": 40.217, |
|
"eval_steps_per_second": 5.067, |
|
"step": 286 |
|
}, |
|
{ |
|
"epoch": 0.46496557310652087, |
|
"grad_norm": 2.021221399307251, |
|
"learning_rate": 0.0001717683695758819, |
|
"loss": 0.2994, |
|
"step": 287 |
|
}, |
|
{ |
|
"epoch": 0.47630619684082626, |
|
"grad_norm": 1.954991340637207, |
|
"learning_rate": 0.00016635157438018983, |
|
"loss": 0.3023, |
|
"step": 294 |
|
}, |
|
{ |
|
"epoch": 0.4844066423653301, |
|
"eval_loss": 0.29642820358276367, |
|
"eval_runtime": 87.2692, |
|
"eval_samples_per_second": 5.729, |
|
"eval_steps_per_second": 0.722, |
|
"step": 299 |
|
}, |
|
{ |
|
"epoch": 0.48764682057513165, |
|
"grad_norm": 1.9946343898773193, |
|
"learning_rate": 0.0001609131054601416, |
|
"loss": 0.2936, |
|
"step": 301 |
|
}, |
|
{ |
|
"epoch": 0.49898744430943703, |
|
"grad_norm": 1.9908640384674072, |
|
"learning_rate": 0.00015546017141067432, |
|
"loss": 0.2827, |
|
"step": 308 |
|
}, |
|
{ |
|
"epoch": 0.5054678007290401, |
|
"eval_loss": 0.2939640283584595, |
|
"eval_runtime": 90.8498, |
|
"eval_samples_per_second": 5.504, |
|
"eval_steps_per_second": 0.693, |
|
"step": 312 |
|
}, |
|
{ |
|
"epoch": 0.5103280680437424, |
|
"grad_norm": 1.9648362398147583, |
|
"learning_rate": 0.00015, |
|
"loss": 0.2795, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 0.5216686917780478, |
|
"grad_norm": 1.9000581502914429, |
|
"learning_rate": 0.0001445398285893257, |
|
"loss": 0.2844, |
|
"step": 322 |
|
}, |
|
{ |
|
"epoch": 0.5265289590927501, |
|
"eval_loss": 0.2863557040691376, |
|
"eval_runtime": 12.4189, |
|
"eval_samples_per_second": 40.261, |
|
"eval_steps_per_second": 5.073, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 0.5330093155123532, |
|
"grad_norm": 1.860827088356018, |
|
"learning_rate": 0.0001390868945398584, |
|
"loss": 0.274, |
|
"step": 329 |
|
}, |
|
{ |
|
"epoch": 0.5443499392466585, |
|
"grad_norm": 2.1385581493377686, |
|
"learning_rate": 0.00013364842561981014, |
|
"loss": 0.2733, |
|
"step": 336 |
|
}, |
|
{ |
|
"epoch": 0.5475901174564601, |
|
"eval_loss": 0.28282585740089417, |
|
"eval_runtime": 92.2187, |
|
"eval_samples_per_second": 5.422, |
|
"eval_steps_per_second": 0.683, |
|
"step": 338 |
|
}, |
|
{ |
|
"epoch": 0.555690562980964, |
|
"grad_norm": 2.0149056911468506, |
|
"learning_rate": 0.00012823163042411807, |
|
"loss": 0.274, |
|
"step": 343 |
|
}, |
|
{ |
|
"epoch": 0.5670311867152693, |
|
"grad_norm": 2.187378406524658, |
|
"learning_rate": 0.0001228436888195789, |
|
"loss": 0.2711, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.56865127582017, |
|
"eval_loss": 0.2776563763618469, |
|
"eval_runtime": 90.611, |
|
"eval_samples_per_second": 5.518, |
|
"eval_steps_per_second": 0.695, |
|
"step": 351 |
|
}, |
|
{ |
|
"epoch": 0.5783718104495748, |
|
"grad_norm": 2.089923143386841, |
|
"learning_rate": 0.00011749174242806152, |
|
"loss": 0.2683, |
|
"step": 357 |
|
}, |
|
{ |
|
"epoch": 0.5897124341838801, |
|
"grad_norm": 1.8779715299606323, |
|
"learning_rate": 0.00011218288516041279, |
|
"loss": 0.2714, |
|
"step": 364 |
|
}, |
|
{ |
|
"epoch": 0.5897124341838801, |
|
"eval_loss": 0.2723066508769989, |
|
"eval_runtime": 12.4698, |
|
"eval_samples_per_second": 40.097, |
|
"eval_steps_per_second": 5.052, |
|
"step": 364 |
|
}, |
|
{ |
|
"epoch": 0.6010530579181855, |
|
"grad_norm": 1.779288411140442, |
|
"learning_rate": 0.00010692415381360349, |
|
"loss": 0.2605, |
|
"step": 371 |
|
}, |
|
{ |
|
"epoch": 0.6107735925475901, |
|
"eval_loss": 0.26519322395324707, |
|
"eval_runtime": 84.674, |
|
"eval_samples_per_second": 5.905, |
|
"eval_steps_per_second": 0.744, |
|
"step": 377 |
|
}, |
|
{ |
|
"epoch": 0.6123936816524909, |
|
"grad_norm": 1.9808777570724487, |
|
"learning_rate": 0.00010172251874357757, |
|
"loss": 0.258, |
|
"step": 378 |
|
}, |
|
{ |
|
"epoch": 0.6237343053867963, |
|
"grad_norm": 1.8243378400802612, |
|
"learning_rate": 9.658487462616794e-05, |
|
"loss": 0.2527, |
|
"step": 385 |
|
}, |
|
{ |
|
"epoch": 0.6318347509113001, |
|
"eval_loss": 0.25875651836395264, |
|
"eval_runtime": 91.6131, |
|
"eval_samples_per_second": 5.458, |
|
"eval_steps_per_second": 0.688, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 0.6350749291211016, |
|
"grad_norm": 1.9190093278884888, |
|
"learning_rate": 9.151803131832493e-05, |
|
"loss": 0.2642, |
|
"step": 392 |
|
}, |
|
{ |
|
"epoch": 0.6464155528554071, |
|
"grad_norm": 1.7457456588745117, |
|
"learning_rate": 8.652870483177049e-05, |
|
"loss": 0.2526, |
|
"step": 399 |
|
}, |
|
{ |
|
"epoch": 0.6528959092750102, |
|
"eval_loss": 0.2529118061065674, |
|
"eval_runtime": 12.4604, |
|
"eval_samples_per_second": 40.127, |
|
"eval_steps_per_second": 5.056, |
|
"step": 403 |
|
}, |
|
{ |
|
"epoch": 0.6577561765897124, |
|
"grad_norm": 1.8457238674163818, |
|
"learning_rate": 8.162350843104291e-05, |
|
"loss": 0.2454, |
|
"step": 406 |
|
}, |
|
{ |
|
"epoch": 0.6690968003240179, |
|
"grad_norm": 2.0805020332336426, |
|
"learning_rate": 7.680894386773072e-05, |
|
"loss": 0.2522, |
|
"step": 413 |
|
}, |
|
{ |
|
"epoch": 0.6739570676387201, |
|
"eval_loss": 0.24899975955486298, |
|
"eval_runtime": 72.2912, |
|
"eval_samples_per_second": 6.916, |
|
"eval_steps_per_second": 0.871, |
|
"step": 416 |
|
}, |
|
{ |
|
"epoch": 0.6804374240583232, |
|
"grad_norm": 1.816220998764038, |
|
"learning_rate": 7.209139276251558e-05, |
|
"loss": 0.2463, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 0.6917780477926286, |
|
"grad_norm": 1.8046435117721558, |
|
"learning_rate": 6.747710814644589e-05, |
|
"loss": 0.242, |
|
"step": 427 |
|
}, |
|
{ |
|
"epoch": 0.6950182260024301, |
|
"eval_loss": 0.2447230964899063, |
|
"eval_runtime": 87.12, |
|
"eval_samples_per_second": 5.739, |
|
"eval_steps_per_second": 0.723, |
|
"step": 429 |
|
}, |
|
{ |
|
"epoch": 0.703118671526934, |
|
"grad_norm": 1.6818225383758545, |
|
"learning_rate": 6.297220617265435e-05, |
|
"loss": 0.2353, |
|
"step": 434 |
|
}, |
|
{ |
|
"epoch": 0.7144592952612394, |
|
"grad_norm": 1.6758326292037964, |
|
"learning_rate": 5.858265800950438e-05, |
|
"loss": 0.2443, |
|
"step": 441 |
|
}, |
|
{ |
|
"epoch": 0.7160793843661402, |
|
"eval_loss": 0.23998361825942993, |
|
"eval_runtime": 12.4908, |
|
"eval_samples_per_second": 40.03, |
|
"eval_steps_per_second": 5.044, |
|
"step": 442 |
|
}, |
|
{ |
|
"epoch": 0.7257999189955447, |
|
"grad_norm": 1.6476861238479614, |
|
"learning_rate": 5.4314281925911634e-05, |
|
"loss": 0.2292, |
|
"step": 448 |
|
}, |
|
{ |
|
"epoch": 0.7371405427298502, |
|
"grad_norm": 1.7244728803634644, |
|
"learning_rate": 5.0172735579330526e-05, |
|
"loss": 0.2273, |
|
"step": 455 |
|
}, |
|
{ |
|
"epoch": 0.7371405427298502, |
|
"eval_loss": 0.23421922326087952, |
|
"eval_runtime": 86.8076, |
|
"eval_samples_per_second": 5.76, |
|
"eval_steps_per_second": 0.726, |
|
"step": 455 |
|
}, |
|
{ |
|
"epoch": 0.7484811664641555, |
|
"grad_norm": 1.8216646909713745, |
|
"learning_rate": 4.616350851662895e-05, |
|
"loss": 0.2288, |
|
"step": 462 |
|
}, |
|
{ |
|
"epoch": 0.7582017010935601, |
|
"eval_loss": 0.23177292943000793, |
|
"eval_runtime": 75.3199, |
|
"eval_samples_per_second": 6.638, |
|
"eval_steps_per_second": 0.836, |
|
"step": 468 |
|
}, |
|
{ |
|
"epoch": 0.759821790198461, |
|
"grad_norm": 1.586669683456421, |
|
"learning_rate": 4.229191489779047e-05, |
|
"loss": 0.2292, |
|
"step": 469 |
|
}, |
|
{ |
|
"epoch": 0.7711624139327663, |
|
"grad_norm": 1.7026342153549194, |
|
"learning_rate": 3.8563086452088506e-05, |
|
"loss": 0.2325, |
|
"step": 476 |
|
}, |
|
{ |
|
"epoch": 0.7792628594572701, |
|
"eval_loss": 0.2292790412902832, |
|
"eval_runtime": 12.4917, |
|
"eval_samples_per_second": 40.026, |
|
"eval_steps_per_second": 5.043, |
|
"step": 481 |
|
}, |
|
{ |
|
"epoch": 0.7825030376670717, |
|
"grad_norm": 1.7042219638824463, |
|
"learning_rate": 3.498196567606959e-05, |
|
"loss": 0.2251, |
|
"step": 483 |
|
}, |
|
{ |
|
"epoch": 0.7938436614013771, |
|
"grad_norm": 1.819270372390747, |
|
"learning_rate": 3.1553299282360966e-05, |
|
"loss": 0.2264, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 0.8003240178209802, |
|
"eval_loss": 0.22619818150997162, |
|
"eval_runtime": 36.3759, |
|
"eval_samples_per_second": 13.745, |
|
"eval_steps_per_second": 1.732, |
|
"step": 494 |
|
}, |
|
{ |
|
"epoch": 0.8051842851356824, |
|
"grad_norm": 1.7286969423294067, |
|
"learning_rate": 2.828163190798644e-05, |
|
"loss": 0.2215, |
|
"step": 497 |
|
}, |
|
{ |
|
"epoch": 0.8165249088699879, |
|
"grad_norm": 1.7360488176345825, |
|
"learning_rate": 2.5171300090530106e-05, |
|
"loss": 0.2127, |
|
"step": 504 |
|
}, |
|
{ |
|
"epoch": 0.8213851761846902, |
|
"eval_loss": 0.2221684604883194, |
|
"eval_runtime": 36.6136, |
|
"eval_samples_per_second": 13.656, |
|
"eval_steps_per_second": 1.721, |
|
"step": 507 |
|
}, |
|
{ |
|
"epoch": 0.8278655326042932, |
|
"grad_norm": 1.7685576677322388, |
|
"learning_rate": 2.2226426520131734e-05, |
|
"loss": 0.2131, |
|
"step": 511 |
|
}, |
|
{ |
|
"epoch": 0.8392061563385986, |
|
"grad_norm": 1.4955885410308838, |
|
"learning_rate": 1.9450914574933725e-05, |
|
"loss": 0.2166, |
|
"step": 518 |
|
}, |
|
{ |
|
"epoch": 0.8424463345484001, |
|
"eval_loss": 0.21989719569683075, |
|
"eval_runtime": 12.5227, |
|
"eval_samples_per_second": 39.927, |
|
"eval_steps_per_second": 5.031, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 0.850546780072904, |
|
"grad_norm": 1.7498217821121216, |
|
"learning_rate": 1.6848443147221828e-05, |
|
"loss": 0.2091, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 0.8618874038072094, |
|
"grad_norm": 1.6159402132034302, |
|
"learning_rate": 1.4422461767118233e-05, |
|
"loss": 0.2145, |
|
"step": 532 |
|
}, |
|
{ |
|
"epoch": 0.8635074929121102, |
|
"eval_loss": 0.21832826733589172, |
|
"eval_runtime": 37.0243, |
|
"eval_samples_per_second": 13.505, |
|
"eval_steps_per_second": 1.702, |
|
"step": 533 |
|
}, |
|
{ |
|
"epoch": 0.8732280275415147, |
|
"grad_norm": 1.4355424642562866, |
|
"learning_rate": 1.2176186030289936e-05, |
|
"loss": 0.2135, |
|
"step": 539 |
|
}, |
|
{ |
|
"epoch": 0.8845686512758202, |
|
"grad_norm": 1.5615161657333374, |
|
"learning_rate": 1.011259333573326e-05, |
|
"loss": 0.2089, |
|
"step": 546 |
|
}, |
|
{ |
|
"epoch": 0.8845686512758202, |
|
"eval_loss": 0.2167486995458603, |
|
"eval_runtime": 36.6749, |
|
"eval_samples_per_second": 13.633, |
|
"eval_steps_per_second": 1.718, |
|
"step": 546 |
|
}, |
|
{ |
|
"epoch": 0.8959092750101255, |
|
"grad_norm": 1.593320608139038, |
|
"learning_rate": 8.234418939283866e-06, |
|
"loss": 0.206, |
|
"step": 553 |
|
}, |
|
{ |
|
"epoch": 0.9056298096395302, |
|
"eval_loss": 0.21581247448921204, |
|
"eval_runtime": 12.5401, |
|
"eval_samples_per_second": 39.872, |
|
"eval_steps_per_second": 5.024, |
|
"step": 559 |
|
}, |
|
{ |
|
"epoch": 0.907249898744431, |
|
"grad_norm": 1.4972048997879028, |
|
"learning_rate": 6.544152328083152e-06, |
|
"loss": 0.2116, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 0.9185905224787363, |
|
"grad_norm": 1.8363592624664307, |
|
"learning_rate": 5.044033920806933e-06, |
|
"loss": 0.2078, |
|
"step": 567 |
|
}, |
|
{ |
|
"epoch": 0.9266909680032401, |
|
"eval_loss": 0.2153376042842865, |
|
"eval_runtime": 36.6671, |
|
"eval_samples_per_second": 13.636, |
|
"eval_steps_per_second": 1.718, |
|
"step": 572 |
|
}, |
|
{ |
|
"epoch": 0.9299311462130417, |
|
"grad_norm": 1.574928879737854, |
|
"learning_rate": 3.7360520980297514e-06, |
|
"loss": 0.2076, |
|
"step": 574 |
|
}, |
|
{ |
|
"epoch": 0.9412717699473471, |
|
"grad_norm": 1.6964603662490845, |
|
"learning_rate": 2.6219405666614402e-06, |
|
"loss": 0.2117, |
|
"step": 581 |
|
}, |
|
{ |
|
"epoch": 0.9477521263669502, |
|
"eval_loss": 0.21471136808395386, |
|
"eval_runtime": 36.6697, |
|
"eval_samples_per_second": 13.635, |
|
"eval_steps_per_second": 1.718, |
|
"step": 585 |
|
}, |
|
{ |
|
"epoch": 0.9526123936816525, |
|
"grad_norm": 1.8814630508422852, |
|
"learning_rate": 1.7031760619491353e-06, |
|
"loss": 0.207, |
|
"step": 588 |
|
}, |
|
{ |
|
"epoch": 0.9639530174159578, |
|
"grad_norm": 1.683011531829834, |
|
"learning_rate": 9.809763900905875e-07, |
|
"loss": 0.2125, |
|
"step": 595 |
|
}, |
|
{ |
|
"epoch": 0.9688132847306602, |
|
"eval_loss": 0.2143588364124298, |
|
"eval_runtime": 12.559, |
|
"eval_samples_per_second": 39.812, |
|
"eval_steps_per_second": 5.016, |
|
"step": 598 |
|
}, |
|
{ |
|
"epoch": 0.9752936411502633, |
|
"grad_norm": 1.555221676826477, |
|
"learning_rate": 4.562988140535073e-07, |
|
"loss": 0.207, |
|
"step": 602 |
|
}, |
|
{ |
|
"epoch": 0.9866342648845686, |
|
"grad_norm": 1.7758488655090332, |
|
"learning_rate": 1.298387847403437e-07, |
|
"loss": 0.2118, |
|
"step": 609 |
|
}, |
|
{ |
|
"epoch": 0.9898744430943702, |
|
"eval_loss": 0.21421405673027039, |
|
"eval_runtime": 36.5871, |
|
"eval_samples_per_second": 13.666, |
|
"eval_steps_per_second": 1.722, |
|
"step": 611 |
|
}, |
|
{ |
|
"epoch": 0.9979748886188741, |
|
"grad_norm": 1.7762000560760498, |
|
"learning_rate": 2.029019180288527e-09, |
|
"loss": 0.2121, |
|
"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": 7.47773295983788e+17, |
|
"train_batch_size": 8, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|