|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.999882513804628, |
|
"global_step": 50000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0, |
|
"learning_rate": 6e-07, |
|
"loss": 8.5434, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"learning_rate": 0.00015, |
|
"loss": 2.6599, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 0.0003, |
|
"loss": 1.2005, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 0.00029998111915108125, |
|
"loss": 1.112, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 0.00029992448135747777, |
|
"loss": 1.0704, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 0.0002998301008774512, |
|
"loss": 1.0454, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 0.00029969800147078263, |
|
"loss": 1.0256, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 0.00029952821639279135, |
|
"loss": 1.0099, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 0.0002993207883859627, |
|
"loss": 1.0, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 0.0002990757696691881, |
|
"loss": 0.9879, |
|
"step": 2250 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 0.00029879322192461925, |
|
"loss": 0.9792, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"eval_loss": 0.9104440212249756, |
|
"eval_runtime": 638.1153, |
|
"eval_samples_per_second": 160.473, |
|
"eval_steps_per_second": 5.015, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 0.0002984732162821399, |
|
"loss": 0.9707, |
|
"step": 2750 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 0.00029811583330145914, |
|
"loss": 0.9635, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 0.0002977211629518312, |
|
"loss": 0.9561, |
|
"step": 3250 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 0.00029728930458940595, |
|
"loss": 0.9507, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 0.0002968203669322168, |
|
"loss": 0.9457, |
|
"step": 3750 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 0.00029631446803281107, |
|
"loss": 0.9406, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 0.00029577173524853123, |
|
"loss": 0.9356, |
|
"step": 4250 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 0.0002951923052094534, |
|
"loss": 0.9314, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 0.00029457632378399127, |
|
"loss": 0.9277, |
|
"step": 4750 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 0.0002939239460421746, |
|
"loss": 0.9238, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"eval_loss": 0.8713345527648926, |
|
"eval_runtime": 575.9949, |
|
"eval_samples_per_second": 177.779, |
|
"eval_steps_per_second": 5.556, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 0.00029323533621661106, |
|
"loss": 0.9198, |
|
"step": 5250 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 0.00029251066766114176, |
|
"loss": 0.9161, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 0.00029175012280720024, |
|
"loss": 0.9131, |
|
"step": 5750 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 0.0002909538931178862, |
|
"loss": 0.9103, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 0.000290122179039766, |
|
"loss": 0.9083, |
|
"step": 6250 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 0.0002892551899524109, |
|
"loss": 0.9048, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 0.0002883531441156872, |
|
"loss": 0.9017, |
|
"step": 6750 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 0.0002874162686148104, |
|
"loss": 0.9001, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 0.00028644479930317775, |
|
"loss": 0.8974, |
|
"step": 7250 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 0.00028543898074299317, |
|
"loss": 0.8953, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"eval_loss": 0.8503240346908569, |
|
"eval_runtime": 567.9032, |
|
"eval_samples_per_second": 180.312, |
|
"eval_steps_per_second": 5.635, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 0.00028439906614370034, |
|
"loss": 0.894, |
|
"step": 7750 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 0.0002833253172982385, |
|
"loss": 0.8916, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 0.0002822180045171373, |
|
"loss": 0.8906, |
|
"step": 8250 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 0.0002810774065604677, |
|
"loss": 0.8881, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 0.0002799038105676658, |
|
"loss": 0.885, |
|
"step": 8750 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 0.0002786975119852465, |
|
"loss": 0.8846, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 0.00027745881449242713, |
|
"loss": 0.8826, |
|
"step": 9250 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 0.0002761880299246772, |
|
"loss": 0.8811, |
|
"step": 9500 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 0.0002748854781952157, |
|
"loss": 0.8788, |
|
"step": 9750 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 0.0002735514872144749, |
|
"loss": 0.8769, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"eval_loss": 0.8355974555015564, |
|
"eval_runtime": 1031.9914, |
|
"eval_samples_per_second": 99.226, |
|
"eval_steps_per_second": 3.101, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 0.0002721863928075503, |
|
"loss": 0.8757, |
|
"step": 10250 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 0.00027079053862965875, |
|
"loss": 0.8736, |
|
"step": 10500 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 0.0002693642760796248, |
|
"loss": 0.8728, |
|
"step": 10750 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 0.00026790796421141813, |
|
"loss": 0.8721, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 0.0002664219696437635, |
|
"loss": 0.8711, |
|
"step": 11250 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 0.00026490666646784665, |
|
"loss": 0.8688, |
|
"step": 11500 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 0.00026336243615313873, |
|
"loss": 0.8671, |
|
"step": 11750 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 0.0002617896674513632, |
|
"loss": 0.8669, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 0.00026018875629862996, |
|
"loss": 0.8651, |
|
"step": 12250 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 0.0002585601057157605, |
|
"loss": 0.8645, |
|
"step": 12500 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"eval_loss": 0.8257074952125549, |
|
"eval_runtime": 305.0402, |
|
"eval_samples_per_second": 335.693, |
|
"eval_steps_per_second": 10.49, |
|
"step": 12500 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 0.00025690412570682946, |
|
"loss": 0.8631, |
|
"step": 12750 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 0.0002552212331559482, |
|
"loss": 0.8615, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 0.0002535118517223168, |
|
"loss": 0.8608, |
|
"step": 13250 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 0.0002517764117335698, |
|
"loss": 0.8603, |
|
"step": 13500 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 0.00025001535007744373, |
|
"loss": 0.8594, |
|
"step": 13750 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 0.00024822911009179276, |
|
"loss": 0.8581, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 0.0002464181414529809, |
|
"loss": 0.8568, |
|
"step": 14250 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 0.00024458290006267833, |
|
"loss": 0.8556, |
|
"step": 14500 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 0.00024272384793309077, |
|
"loss": 0.8557, |
|
"step": 14750 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 0.00024084145307064997, |
|
"loss": 0.854, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"eval_loss": 0.8169683218002319, |
|
"eval_runtime": 299.799, |
|
"eval_samples_per_second": 341.562, |
|
"eval_steps_per_second": 10.674, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 0.00023893618935819607, |
|
"loss": 0.8538, |
|
"step": 15250 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 0.0002370085364356797, |
|
"loss": 0.8524, |
|
"step": 15500 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 0.00023505897957941556, |
|
"loss": 0.8519, |
|
"step": 15750 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 0.00023308800957991653, |
|
"loss": 0.8515, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 0.00023109612261833963, |
|
"loss": 0.8506, |
|
"step": 16250 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 0.00022908382014157533, |
|
"loss": 0.8492, |
|
"step": 16500 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 0.00022705160873601096, |
|
"loss": 0.8477, |
|
"step": 16750 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"learning_rate": 0.000225, |
|
"loss": 0.8477, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"learning_rate": 0.00022292951041507028, |
|
"loss": 0.8475, |
|
"step": 17250 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 0.0002208406612159024, |
|
"loss": 0.8457, |
|
"step": 17500 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"eval_loss": 0.8102055191993713, |
|
"eval_runtime": 297.9674, |
|
"eval_samples_per_second": 343.662, |
|
"eval_steps_per_second": 10.739, |
|
"step": 17500 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 0.00021873397825911153, |
|
"loss": 0.8457, |
|
"step": 17750 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"learning_rate": 0.0002166099918908661, |
|
"loss": 0.8447, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"learning_rate": 0.00021446923681337575, |
|
"loss": 0.8433, |
|
"step": 18250 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"learning_rate": 0.00021231225195028297, |
|
"loss": 0.8435, |
|
"step": 18500 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"learning_rate": 0.00021013958031099205, |
|
"loss": 0.8437, |
|
"step": 18750 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 0.00020795176885396926, |
|
"loss": 0.8424, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 0.0002057493683490491, |
|
"loss": 0.8411, |
|
"step": 19250 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 0.00020353293323878074, |
|
"loss": 0.8406, |
|
"step": 19500 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 0.00020130302149885031, |
|
"loss": 0.8407, |
|
"step": 19750 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"learning_rate": 0.00019906019449761325, |
|
"loss": 0.8396, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"eval_loss": 0.8049243092536926, |
|
"eval_runtime": 297.8022, |
|
"eval_samples_per_second": 343.852, |
|
"eval_steps_per_second": 10.745, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"learning_rate": 0.00019680501685477304, |
|
"loss": 0.8388, |
|
"step": 20250 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"learning_rate": 0.00019453805629924124, |
|
"loss": 0.8382, |
|
"step": 20500 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"learning_rate": 0.00019225988352621445, |
|
"loss": 0.8379, |
|
"step": 20750 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 0.0001899710720535052, |
|
"loss": 0.8372, |
|
"step": 21000 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 0.00018767219807716185, |
|
"loss": 0.8366, |
|
"step": 21250 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 0.0001853638403264141, |
|
"loss": 0.8365, |
|
"step": 21500 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 0.0001830465799179811, |
|
"loss": 0.835, |
|
"step": 21750 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"learning_rate": 0.0001807210002097786, |
|
"loss": 0.8342, |
|
"step": 22000 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"learning_rate": 0.0001783876866540615, |
|
"loss": 0.8348, |
|
"step": 22250 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 0.00017604722665003956, |
|
"loss": 0.8335, |
|
"step": 22500 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"eval_loss": 0.799386739730835, |
|
"eval_runtime": 297.9025, |
|
"eval_samples_per_second": 343.737, |
|
"eval_steps_per_second": 10.742, |
|
"step": 22500 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 0.00017370020939600248, |
|
"loss": 0.8332, |
|
"step": 22750 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"learning_rate": 0.00017134722574099276, |
|
"loss": 0.8323, |
|
"step": 23000 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"learning_rate": 0.00016898886803606237, |
|
"loss": 0.8322, |
|
"step": 23250 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"learning_rate": 0.00016662572998515164, |
|
"loss": 0.8323, |
|
"step": 23500 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"learning_rate": 0.00016425840649562736, |
|
"loss": 0.8314, |
|
"step": 23750 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 0.00016188749352851825, |
|
"loss": 0.8308, |
|
"step": 24000 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 0.00015951358794848465, |
|
"loss": 0.8298, |
|
"step": 24250 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 0.00015713728737356137, |
|
"loss": 0.8302, |
|
"step": 24500 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 0.00015475919002471016, |
|
"loss": 0.83, |
|
"step": 24750 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"learning_rate": 0.00015237989457522118, |
|
"loss": 0.8288, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"eval_loss": 0.7951927781105042, |
|
"eval_runtime": 297.9768, |
|
"eval_samples_per_second": 343.651, |
|
"eval_steps_per_second": 10.739, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"learning_rate": 0.00015, |
|
"loss": 0.8289, |
|
"step": 25250 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"learning_rate": 0.0001476201054247788, |
|
"loss": 0.8279, |
|
"step": 25500 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"learning_rate": 0.00014524080997528987, |
|
"loss": 0.8276, |
|
"step": 25750 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"learning_rate": 0.00014286271262643866, |
|
"loss": 0.8276, |
|
"step": 26000 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"learning_rate": 0.00014048641205151533, |
|
"loss": 0.8268, |
|
"step": 26250 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"learning_rate": 0.0001381125064714817, |
|
"loss": 0.8261, |
|
"step": 26500 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"learning_rate": 0.00013574159350437261, |
|
"loss": 0.826, |
|
"step": 26750 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 0.00013337427001484836, |
|
"loss": 0.8254, |
|
"step": 27000 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 0.00013101113196393758, |
|
"loss": 0.8253, |
|
"step": 27250 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"learning_rate": 0.00012865277425900724, |
|
"loss": 0.8249, |
|
"step": 27500 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"eval_loss": 0.7913097143173218, |
|
"eval_runtime": 298.0125, |
|
"eval_samples_per_second": 343.61, |
|
"eval_steps_per_second": 10.738, |
|
"step": 27500 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"learning_rate": 0.0001262997906039975, |
|
"loss": 0.8247, |
|
"step": 27750 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"learning_rate": 0.00012395277334996044, |
|
"loss": 0.8239, |
|
"step": 28000 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"learning_rate": 0.00012161231334593851, |
|
"loss": 0.8241, |
|
"step": 28250 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 0.0001192789997902214, |
|
"loss": 0.8237, |
|
"step": 28500 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 0.00011695342008201888, |
|
"loss": 0.8235, |
|
"step": 28750 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"learning_rate": 0.00011463615967358588, |
|
"loss": 0.8229, |
|
"step": 29000 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"learning_rate": 0.00011232780192283812, |
|
"loss": 0.8221, |
|
"step": 29250 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"learning_rate": 0.00011002892794649476, |
|
"loss": 0.8214, |
|
"step": 29500 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"learning_rate": 0.00010774011647378553, |
|
"loss": 0.8213, |
|
"step": 29750 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"learning_rate": 0.00010546194370075881, |
|
"loss": 0.8209, |
|
"step": 30000 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"eval_loss": 0.788076639175415, |
|
"eval_runtime": 298.3169, |
|
"eval_samples_per_second": 343.259, |
|
"eval_steps_per_second": 10.727, |
|
"step": 30000 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"learning_rate": 0.00010319498314522693, |
|
"loss": 0.8212, |
|
"step": 30250 |
|
}, |
|
{ |
|
"epoch": 0.61, |
|
"learning_rate": 0.00010093980550238675, |
|
"loss": 0.8201, |
|
"step": 30500 |
|
}, |
|
{ |
|
"epoch": 0.61, |
|
"learning_rate": 9.869697850114969e-05, |
|
"loss": 0.8212, |
|
"step": 30750 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"learning_rate": 9.646706676121923e-05, |
|
"loss": 0.8207, |
|
"step": 31000 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"learning_rate": 9.425063165095088e-05, |
|
"loss": 0.8197, |
|
"step": 31250 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"learning_rate": 9.204823114603068e-05, |
|
"loss": 0.8196, |
|
"step": 31500 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"learning_rate": 8.986041968900796e-05, |
|
"loss": 0.819, |
|
"step": 31750 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"learning_rate": 8.768774804971705e-05, |
|
"loss": 0.8194, |
|
"step": 32000 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"learning_rate": 8.553076318662425e-05, |
|
"loss": 0.8189, |
|
"step": 32250 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"learning_rate": 8.339000810913386e-05, |
|
"loss": 0.8187, |
|
"step": 32500 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"eval_loss": 0.7861111164093018, |
|
"eval_runtime": 284.3275, |
|
"eval_samples_per_second": 360.148, |
|
"eval_steps_per_second": 5.627, |
|
"step": 32500 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"learning_rate": 8.126602174088843e-05, |
|
"loss": 0.8187, |
|
"step": 32750 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 7.915933878409761e-05, |
|
"loss": 0.8175, |
|
"step": 33000 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 7.707048958492972e-05, |
|
"loss": 0.8171, |
|
"step": 33250 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"learning_rate": 7.500000000000002e-05, |
|
"loss": 0.8175, |
|
"step": 33500 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"learning_rate": 7.294839126398908e-05, |
|
"loss": 0.8173, |
|
"step": 33750 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"learning_rate": 7.091617985842462e-05, |
|
"loss": 0.8166, |
|
"step": 34000 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"learning_rate": 6.890387738166041e-05, |
|
"loss": 0.8164, |
|
"step": 34250 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"learning_rate": 6.691199042008345e-05, |
|
"loss": 0.8164, |
|
"step": 34500 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"learning_rate": 6.49410204205844e-05, |
|
"loss": 0.8163, |
|
"step": 34750 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 6.299146356432029e-05, |
|
"loss": 0.8162, |
|
"step": 35000 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"eval_loss": 0.783399224281311, |
|
"eval_runtime": 284.3579, |
|
"eval_samples_per_second": 360.11, |
|
"eval_steps_per_second": 5.627, |
|
"step": 35000 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 6.106381064180395e-05, |
|
"loss": 0.8159, |
|
"step": 35250 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"learning_rate": 5.915854692935002e-05, |
|
"loss": 0.8158, |
|
"step": 35500 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"learning_rate": 5.72761520669092e-05, |
|
"loss": 0.8155, |
|
"step": 35750 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"learning_rate": 5.541709993732167e-05, |
|
"loss": 0.8153, |
|
"step": 36000 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"learning_rate": 5.358185854701909e-05, |
|
"loss": 0.8154, |
|
"step": 36250 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"learning_rate": 5.1770889908207245e-05, |
|
"loss": 0.8151, |
|
"step": 36500 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"learning_rate": 4.998464992255627e-05, |
|
"loss": 0.8146, |
|
"step": 36750 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"learning_rate": 4.8223588266430186e-05, |
|
"loss": 0.814, |
|
"step": 37000 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"learning_rate": 4.648814827768322e-05, |
|
"loss": 0.8143, |
|
"step": 37250 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"learning_rate": 4.477876684405179e-05, |
|
"loss": 0.8139, |
|
"step": 37500 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"eval_loss": 0.7822485566139221, |
|
"eval_runtime": 284.2914, |
|
"eval_samples_per_second": 360.194, |
|
"eval_steps_per_second": 5.628, |
|
"step": 37500 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"learning_rate": 4.309587429317061e-05, |
|
"loss": 0.8143, |
|
"step": 37750 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 4.143989428423947e-05, |
|
"loss": 0.8142, |
|
"step": 38000 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 3.981124370137001e-05, |
|
"loss": 0.8143, |
|
"step": 38250 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"learning_rate": 3.8210332548636796e-05, |
|
"loss": 0.8137, |
|
"step": 38500 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"learning_rate": 3.663756384686127e-05, |
|
"loss": 0.8135, |
|
"step": 38750 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"learning_rate": 3.509333353215331e-05, |
|
"loss": 0.8133, |
|
"step": 39000 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"learning_rate": 3.3578030356236455e-05, |
|
"loss": 0.8133, |
|
"step": 39250 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"learning_rate": 3.209203578858191e-05, |
|
"loss": 0.8124, |
|
"step": 39500 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"learning_rate": 3.0635723920375164e-05, |
|
"loss": 0.8121, |
|
"step": 39750 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"learning_rate": 2.9209461370341204e-05, |
|
"loss": 0.8122, |
|
"step": 40000 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"eval_loss": 0.7810642719268799, |
|
"eval_runtime": 284.3798, |
|
"eval_samples_per_second": 360.082, |
|
"eval_steps_per_second": 5.626, |
|
"step": 40000 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"learning_rate": 2.781360719244964e-05, |
|
"loss": 0.8127, |
|
"step": 40250 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"learning_rate": 2.6448512785525093e-05, |
|
"loss": 0.8138, |
|
"step": 40500 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"learning_rate": 2.5114521804784305e-05, |
|
"loss": 0.8129, |
|
"step": 40750 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 2.38119700753228e-05, |
|
"loss": 0.8121, |
|
"step": 41000 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 2.2541185507572858e-05, |
|
"loss": 0.8126, |
|
"step": 41250 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"learning_rate": 2.130248801475344e-05, |
|
"loss": 0.8121, |
|
"step": 41500 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"learning_rate": 2.009618943233419e-05, |
|
"loss": 0.8127, |
|
"step": 41750 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"learning_rate": 1.892259343953226e-05, |
|
"loss": 0.8131, |
|
"step": 42000 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"learning_rate": 1.7781995482862705e-05, |
|
"loss": 0.8124, |
|
"step": 42250 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"learning_rate": 1.6674682701761493e-05, |
|
"loss": 0.8122, |
|
"step": 42500 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"eval_loss": 0.7801794409751892, |
|
"eval_runtime": 284.3702, |
|
"eval_samples_per_second": 360.094, |
|
"eval_steps_per_second": 5.626, |
|
"step": 42500 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"learning_rate": 1.5600933856299635e-05, |
|
"loss": 0.8121, |
|
"step": 42750 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"learning_rate": 1.4561019257006839e-05, |
|
"loss": 0.8114, |
|
"step": 43000 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"learning_rate": 1.3555200696822232e-05, |
|
"loss": 0.812, |
|
"step": 43250 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"learning_rate": 1.258373138518956e-05, |
|
"loss": 0.8116, |
|
"step": 43500 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"learning_rate": 1.164685588431281e-05, |
|
"loss": 0.8112, |
|
"step": 43750 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"learning_rate": 1.0744810047589115e-05, |
|
"loss": 0.8116, |
|
"step": 44000 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"learning_rate": 9.877820960234002e-06, |
|
"loss": 0.8116, |
|
"step": 44250 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"learning_rate": 9.046106882113751e-06, |
|
"loss": 0.811, |
|
"step": 44500 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"learning_rate": 8.249877192799731e-06, |
|
"loss": 0.8109, |
|
"step": 44750 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"learning_rate": 7.489332338858201e-06, |
|
"loss": 0.8113, |
|
"step": 45000 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"eval_loss": 0.7795301675796509, |
|
"eval_runtime": 284.3564, |
|
"eval_samples_per_second": 360.111, |
|
"eval_steps_per_second": 5.627, |
|
"step": 45000 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"learning_rate": 6.764663783388918e-06, |
|
"loss": 0.8118, |
|
"step": 45250 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 6.076053957825411e-06, |
|
"loss": 0.8112, |
|
"step": 45500 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 5.4236762160086935e-06, |
|
"loss": 0.8119, |
|
"step": 45750 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"learning_rate": 4.807694790546562e-06, |
|
"loss": 0.811, |
|
"step": 46000 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"learning_rate": 4.228264751468752e-06, |
|
"loss": 0.8116, |
|
"step": 46250 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"learning_rate": 3.6855319671889427e-06, |
|
"loss": 0.8111, |
|
"step": 46500 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"learning_rate": 3.179633067783205e-06, |
|
"loss": 0.8119, |
|
"step": 46750 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"learning_rate": 2.710695410593994e-06, |
|
"loss": 0.8112, |
|
"step": 47000 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"learning_rate": 2.2788370481687965e-06, |
|
"loss": 0.8122, |
|
"step": 47250 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"learning_rate": 1.8841666985408566e-06, |
|
"loss": 0.8116, |
|
"step": 47500 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"eval_loss": 0.779478907585144, |
|
"eval_runtime": 284.4432, |
|
"eval_samples_per_second": 360.002, |
|
"eval_steps_per_second": 5.625, |
|
"step": 47500 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"learning_rate": 1.5267837178600972e-06, |
|
"loss": 0.8116, |
|
"step": 47750 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"learning_rate": 1.2067780753806988e-06, |
|
"loss": 0.8113, |
|
"step": 48000 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"learning_rate": 9.242303308118815e-07, |
|
"loss": 0.8107, |
|
"step": 48250 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 6.792116140373116e-07, |
|
"loss": 0.8116, |
|
"step": 48500 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 4.717836072086589e-07, |
|
"loss": 0.8111, |
|
"step": 48750 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 3.01998529217351e-07, |
|
"loss": 0.8108, |
|
"step": 49000 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 1.6989912254880556e-07, |
|
"loss": 0.8112, |
|
"step": 49250 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"learning_rate": 7.551864252223761e-08, |
|
"loss": 0.8112, |
|
"step": 49500 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"learning_rate": 1.8880848918739756e-08, |
|
"loss": 0.8116, |
|
"step": 49750 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 0.0, |
|
"loss": 0.8104, |
|
"step": 50000 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 0.7793822288513184, |
|
"eval_runtime": 590.5864, |
|
"eval_samples_per_second": 173.387, |
|
"eval_steps_per_second": 5.418, |
|
"step": 50000 |
|
} |
|
], |
|
"max_steps": 50000, |
|
"num_train_epochs": 1, |
|
"total_flos": 4.5371203009812365e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|