| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 478, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "eta": 0.004999999888241291, | |
| "learning_rate": 1.0416666666666666e-08, | |
| "logits/chosen": -2.745321273803711, | |
| "logits/rejected": -2.661250352859497, | |
| "logps/chosen": -321.0613098144531, | |
| "logps/rejected": -271.1681823730469, | |
| "loss": 0.7079, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 1.0416666666666667e-07, | |
| "logits/chosen": -2.781073570251465, | |
| "logits/rejected": -2.7723324298858643, | |
| "logps/chosen": -242.78675842285156, | |
| "logps/rejected": -208.898193359375, | |
| "loss": 0.7064, | |
| "rewards/accuracies": 0.5069444179534912, | |
| "rewards/chosen": 0.00046142542851157486, | |
| "rewards/margins": 0.0004825991054531187, | |
| "rewards/rejected": -2.1173778804950416e-05, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 2.0833333333333333e-07, | |
| "logits/chosen": -2.7764596939086914, | |
| "logits/rejected": -2.76332426071167, | |
| "logps/chosen": -268.7160339355469, | |
| "logps/rejected": -246.1806182861328, | |
| "loss": 0.7059, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": 0.0012417829129844904, | |
| "rewards/margins": 0.0018564596539363265, | |
| "rewards/rejected": -0.0006146768573671579, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 3.1249999999999997e-07, | |
| "logits/chosen": -2.829183578491211, | |
| "logits/rejected": -2.789580821990967, | |
| "logps/chosen": -279.50506591796875, | |
| "logps/rejected": -272.59332275390625, | |
| "loss": 0.703, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": 0.001455739140510559, | |
| "rewards/margins": 0.005600649863481522, | |
| "rewards/rejected": -0.0041449107229709625, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.1666666666666667e-07, | |
| "logits/chosen": -2.7342190742492676, | |
| "logits/rejected": -2.7025771141052246, | |
| "logps/chosen": -287.4779968261719, | |
| "logps/rejected": -262.9581298828125, | |
| "loss": 0.6913, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": 0.00810016505420208, | |
| "rewards/margins": 0.03219769150018692, | |
| "rewards/rejected": -0.02409752830862999, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.999733114418725e-07, | |
| "logits/chosen": -2.70398211479187, | |
| "logits/rejected": -2.678131580352783, | |
| "logps/chosen": -306.2682189941406, | |
| "logps/rejected": -304.858642578125, | |
| "loss": 0.6757, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -0.010384158231317997, | |
| "rewards/margins": 0.06137201189994812, | |
| "rewards/rejected": -0.07175617665052414, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.990398100856366e-07, | |
| "logits/chosen": -2.6752824783325195, | |
| "logits/rejected": -2.6444764137268066, | |
| "logps/chosen": -260.00921630859375, | |
| "logps/rejected": -241.9503173828125, | |
| "loss": 0.66, | |
| "rewards/accuracies": 0.675000011920929, | |
| "rewards/chosen": -0.07034246623516083, | |
| "rewards/margins": 0.10934233665466309, | |
| "rewards/rejected": -0.17968478798866272, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.967775735898179e-07, | |
| "logits/chosen": -2.7276742458343506, | |
| "logits/rejected": -2.7039637565612793, | |
| "logps/chosen": -294.48309326171875, | |
| "logps/rejected": -287.7945251464844, | |
| "loss": 0.6352, | |
| "rewards/accuracies": 0.65625, | |
| "rewards/chosen": -0.21306617558002472, | |
| "rewards/margins": 0.20508523285388947, | |
| "rewards/rejected": -0.4181514382362366, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.931986719649298e-07, | |
| "logits/chosen": -2.7324066162109375, | |
| "logits/rejected": -2.7214457988739014, | |
| "logps/chosen": -303.2037048339844, | |
| "logps/rejected": -328.5882873535156, | |
| "loss": 0.6282, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": -0.38951486349105835, | |
| "rewards/margins": 0.1954750418663025, | |
| "rewards/rejected": -0.5849899053573608, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.883222001996351e-07, | |
| "logits/chosen": -2.732830286026001, | |
| "logits/rejected": -2.6980254650115967, | |
| "logps/chosen": -297.2022399902344, | |
| "logps/rejected": -315.86553955078125, | |
| "loss": 0.5956, | |
| "rewards/accuracies": 0.6625000238418579, | |
| "rewards/chosen": -0.3561047911643982, | |
| "rewards/margins": 0.31715089082717896, | |
| "rewards/rejected": -0.6732556819915771, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.821741763807186e-07, | |
| "logits/chosen": -2.7799582481384277, | |
| "logits/rejected": -2.734489679336548, | |
| "logps/chosen": -322.6822204589844, | |
| "logps/rejected": -346.29632568359375, | |
| "loss": 0.5808, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": -0.4515351355075836, | |
| "rewards/margins": 0.430880069732666, | |
| "rewards/rejected": -0.8824151158332825, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.747874028753375e-07, | |
| "logits/chosen": -2.737361431121826, | |
| "logits/rejected": -2.7130186557769775, | |
| "logps/chosen": -354.4649353027344, | |
| "logps/rejected": -358.72894287109375, | |
| "loss": 0.5836, | |
| "rewards/accuracies": 0.731249988079071, | |
| "rewards/chosen": -0.42983478307724, | |
| "rewards/margins": 0.48764386773109436, | |
| "rewards/rejected": -0.9174786806106567, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.662012913161997e-07, | |
| "logits/chosen": -2.5539088249206543, | |
| "logits/rejected": -2.5288946628570557, | |
| "logps/chosen": -332.5165710449219, | |
| "logps/rejected": -339.154296875, | |
| "loss": 0.5704, | |
| "rewards/accuracies": 0.75, | |
| "rewards/chosen": -0.5978292226791382, | |
| "rewards/margins": 0.5392987728118896, | |
| "rewards/rejected": -1.1371279954910278, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.5646165232345103e-07, | |
| "logits/chosen": -2.4799418449401855, | |
| "logits/rejected": -2.463563919067383, | |
| "logps/chosen": -309.03570556640625, | |
| "logps/rejected": -393.3725280761719, | |
| "loss": 0.5759, | |
| "rewards/accuracies": 0.7437499761581421, | |
| "rewards/chosen": -0.5238430500030518, | |
| "rewards/margins": 0.5551779866218567, | |
| "rewards/rejected": -1.0790210962295532, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.456204510851956e-07, | |
| "logits/chosen": -2.4601731300354004, | |
| "logits/rejected": -2.4274346828460693, | |
| "logps/chosen": -369.95086669921875, | |
| "logps/rejected": -402.4490661621094, | |
| "loss": 0.5515, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": -0.8267312049865723, | |
| "rewards/margins": 0.6819665431976318, | |
| "rewards/rejected": -1.508697748184204, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.337355301007335e-07, | |
| "logits/chosen": -2.4591126441955566, | |
| "logits/rejected": -2.4192230701446533, | |
| "logps/chosen": -338.8128356933594, | |
| "logps/rejected": -341.96185302734375, | |
| "loss": 0.5644, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": -0.55790776014328, | |
| "rewards/margins": 0.5498847961425781, | |
| "rewards/rejected": -1.107792615890503, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.2087030056579986e-07, | |
| "logits/chosen": -2.324352979660034, | |
| "logits/rejected": -2.297412872314453, | |
| "logps/chosen": -333.8399963378906, | |
| "logps/rejected": -388.5691833496094, | |
| "loss": 0.5386, | |
| "rewards/accuracies": 0.7749999761581421, | |
| "rewards/chosen": -0.5367451906204224, | |
| "rewards/margins": 0.6883090138435364, | |
| "rewards/rejected": -1.225054144859314, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.070934040463998e-07, | |
| "logits/chosen": -2.25808048248291, | |
| "logits/rejected": -2.191967010498047, | |
| "logps/chosen": -348.13031005859375, | |
| "logps/rejected": -380.72906494140625, | |
| "loss": 0.5827, | |
| "rewards/accuracies": 0.6812499761581421, | |
| "rewards/chosen": -1.0580767393112183, | |
| "rewards/margins": 0.4786924421787262, | |
| "rewards/rejected": -1.536769151687622, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 3.9247834624635404e-07, | |
| "logits/chosen": -2.3127024173736572, | |
| "logits/rejected": -2.320852279663086, | |
| "logps/chosen": -367.42950439453125, | |
| "logps/rejected": -426.7998046875, | |
| "loss": 0.5569, | |
| "rewards/accuracies": 0.6875, | |
| "rewards/chosen": -0.7815980315208435, | |
| "rewards/margins": 0.7305761575698853, | |
| "rewards/rejected": -1.512174129486084, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 3.7710310482256523e-07, | |
| "logits/chosen": -2.4002134799957275, | |
| "logits/rejected": -2.3388237953186035, | |
| "logps/chosen": -328.8986511230469, | |
| "logps/rejected": -381.8426208496094, | |
| "loss": 0.5436, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": -0.5217610597610474, | |
| "rewards/margins": 0.6428496241569519, | |
| "rewards/rejected": -1.1646106243133545, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 3.610497133404795e-07, | |
| "logits/chosen": -2.2329065799713135, | |
| "logits/rejected": -2.1769356727600098, | |
| "logps/chosen": -364.03607177734375, | |
| "logps/rejected": -379.78289794921875, | |
| "loss": 0.5399, | |
| "rewards/accuracies": 0.643750011920929, | |
| "rewards/chosen": -0.8814976811408997, | |
| "rewards/margins": 0.5185674428939819, | |
| "rewards/rejected": -1.4000650644302368, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 3.4440382358952115e-07, | |
| "logits/chosen": -2.2697091102600098, | |
| "logits/rejected": -2.2113330364227295, | |
| "logps/chosen": -344.99200439453125, | |
| "logps/rejected": -368.27581787109375, | |
| "loss": 0.554, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": -0.7441986799240112, | |
| "rewards/margins": 0.5876610279083252, | |
| "rewards/rejected": -1.331859827041626, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 3.272542485937368e-07, | |
| "logits/chosen": -2.227048873901367, | |
| "logits/rejected": -2.2157351970672607, | |
| "logps/chosen": -324.1903991699219, | |
| "logps/rejected": -370.2602844238281, | |
| "loss": 0.5538, | |
| "rewards/accuracies": 0.7875000238418579, | |
| "rewards/chosen": -0.5380795001983643, | |
| "rewards/margins": 0.7564869523048401, | |
| "rewards/rejected": -1.2945663928985596, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 3.096924887558854e-07, | |
| "logits/chosen": -2.1815030574798584, | |
| "logits/rejected": -2.1446516513824463, | |
| "logps/chosen": -329.86566162109375, | |
| "logps/rejected": -409.99267578125, | |
| "loss": 0.5485, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": -0.6506673097610474, | |
| "rewards/margins": 0.7174521684646606, | |
| "rewards/rejected": -1.3681195974349976, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 2.9181224366319943e-07, | |
| "logits/chosen": -2.1926796436309814, | |
| "logits/rejected": -2.133800745010376, | |
| "logps/chosen": -328.4383850097656, | |
| "logps/rejected": -380.1564025878906, | |
| "loss": 0.5478, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": -0.8131483793258667, | |
| "rewards/margins": 0.6653987169265747, | |
| "rewards/rejected": -1.4785473346710205, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 2.7370891215954565e-07, | |
| "logits/chosen": -2.198378324508667, | |
| "logits/rejected": -2.148686647415161, | |
| "logps/chosen": -343.18426513671875, | |
| "logps/rejected": -405.7811584472656, | |
| "loss": 0.5379, | |
| "rewards/accuracies": 0.699999988079071, | |
| "rewards/chosen": -0.8210641741752625, | |
| "rewards/margins": 0.6968287229537964, | |
| "rewards/rejected": -1.5178929567337036, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 2.55479083351317e-07, | |
| "logits/chosen": -2.1465113162994385, | |
| "logits/rejected": -2.093456745147705, | |
| "logps/chosen": -347.0015869140625, | |
| "logps/rejected": -396.0328063964844, | |
| "loss": 0.5184, | |
| "rewards/accuracies": 0.8062499761581421, | |
| "rewards/chosen": -0.7407617568969727, | |
| "rewards/margins": 0.8684328198432922, | |
| "rewards/rejected": -1.6091945171356201, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 2.3722002126275822e-07, | |
| "logits/chosen": -2.1429693698883057, | |
| "logits/rejected": -2.0827112197875977, | |
| "logps/chosen": -358.7803649902344, | |
| "logps/rejected": -440.46990966796875, | |
| "loss": 0.527, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -0.9018821716308594, | |
| "rewards/margins": 0.9309617877006531, | |
| "rewards/rejected": -1.8328437805175781, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 2.19029145890313e-07, | |
| "logits/chosen": -2.215878486633301, | |
| "logits/rejected": -2.1548526287078857, | |
| "logps/chosen": -337.1029357910156, | |
| "logps/rejected": -380.2957763671875, | |
| "loss": 0.5418, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": -0.6890209913253784, | |
| "rewards/margins": 0.7738422155380249, | |
| "rewards/rejected": -1.4628633260726929, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 2.0100351342479216e-07, | |
| "logits/chosen": -2.1938037872314453, | |
| "logits/rejected": -2.177739381790161, | |
| "logps/chosen": -353.51544189453125, | |
| "logps/rejected": -412.64306640625, | |
| "loss": 0.5398, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -0.726887583732605, | |
| "rewards/margins": 0.7154626250267029, | |
| "rewards/rejected": -1.442350149154663, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 1.8323929841460178e-07, | |
| "logits/chosen": -2.1435744762420654, | |
| "logits/rejected": -2.082127332687378, | |
| "logps/chosen": -368.35906982421875, | |
| "logps/rejected": -436.9691467285156, | |
| "loss": 0.5324, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": -0.8049615025520325, | |
| "rewards/margins": 0.6589316725730896, | |
| "rewards/rejected": -1.4638930559158325, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 1.6583128063291573e-07, | |
| "logits/chosen": -2.188692092895508, | |
| "logits/rejected": -2.1064612865448, | |
| "logps/chosen": -354.7235412597656, | |
| "logps/rejected": -376.1325378417969, | |
| "loss": 0.5187, | |
| "rewards/accuracies": 0.731249988079071, | |
| "rewards/chosen": -0.8574191331863403, | |
| "rewards/margins": 0.742445707321167, | |
| "rewards/rejected": -1.5998647212982178, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 1.488723393865766e-07, | |
| "logits/chosen": -2.113239288330078, | |
| "logits/rejected": -2.0523862838745117, | |
| "logps/chosen": -317.50787353515625, | |
| "logps/rejected": -413.80426025390625, | |
| "loss": 0.522, | |
| "rewards/accuracies": 0.793749988079071, | |
| "rewards/chosen": -0.7432131171226501, | |
| "rewards/margins": 1.0518220663070679, | |
| "rewards/rejected": -1.7950351238250732, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 1.3245295796480788e-07, | |
| "logits/chosen": -2.2147035598754883, | |
| "logits/rejected": -2.1897997856140137, | |
| "logps/chosen": -375.3721618652344, | |
| "logps/rejected": -414.7781677246094, | |
| "loss": 0.5318, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -0.9661127328872681, | |
| "rewards/margins": 0.6892365217208862, | |
| "rewards/rejected": -1.6553493738174438, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 1.1666074087171627e-07, | |
| "logits/chosen": -2.1029276847839355, | |
| "logits/rejected": -2.0935397148132324, | |
| "logps/chosen": -336.9126892089844, | |
| "logps/rejected": -391.1408386230469, | |
| "loss": 0.5351, | |
| "rewards/accuracies": 0.706250011920929, | |
| "rewards/chosen": -0.8055974841117859, | |
| "rewards/margins": 0.6507540941238403, | |
| "rewards/rejected": -1.4563515186309814, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 1.0157994641835734e-07, | |
| "logits/chosen": -2.0941309928894043, | |
| "logits/rejected": -2.0219178199768066, | |
| "logps/chosen": -322.93145751953125, | |
| "logps/rejected": -389.5673828125, | |
| "loss": 0.5032, | |
| "rewards/accuracies": 0.7875000238418579, | |
| "rewards/chosen": -0.7798911333084106, | |
| "rewards/margins": 0.814423680305481, | |
| "rewards/rejected": -1.5943149328231812, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 8.729103716819111e-08, | |
| "logits/chosen": -2.1901535987854004, | |
| "logits/rejected": -2.1034839153289795, | |
| "logps/chosen": -386.5938720703125, | |
| "logps/rejected": -423.3191833496094, | |
| "loss": 0.5109, | |
| "rewards/accuracies": 0.7562500238418579, | |
| "rewards/chosen": -0.7419500946998596, | |
| "rewards/margins": 0.9194036722183228, | |
| "rewards/rejected": -1.6613538265228271, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 7.387025063449081e-08, | |
| "logits/chosen": -2.1681642532348633, | |
| "logits/rejected": -2.0902836322784424, | |
| "logps/chosen": -379.05096435546875, | |
| "logps/rejected": -418.29620361328125, | |
| "loss": 0.5313, | |
| "rewards/accuracies": 0.824999988079071, | |
| "rewards/chosen": -0.7454948425292969, | |
| "rewards/margins": 0.9174238443374634, | |
| "rewards/rejected": -1.6629188060760498, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 6.138919252022435e-08, | |
| "logits/chosen": -2.127246618270874, | |
| "logits/rejected": -2.0689470767974854, | |
| "logps/chosen": -356.09906005859375, | |
| "logps/rejected": -440.26788330078125, | |
| "loss": 0.5075, | |
| "rewards/accuracies": 0.7749999761581421, | |
| "rewards/chosen": -0.8266879320144653, | |
| "rewards/margins": 0.8445302844047546, | |
| "rewards/rejected": -1.6712181568145752, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.991445467064689e-08, | |
| "logits/chosen": -2.1151671409606934, | |
| "logits/rejected": -2.071991443634033, | |
| "logps/chosen": -353.2530822753906, | |
| "logps/rejected": -428.5403747558594, | |
| "loss": 0.5323, | |
| "rewards/accuracies": 0.7250000238418579, | |
| "rewards/chosen": -0.8702594637870789, | |
| "rewards/margins": 0.7266938090324402, | |
| "rewards/rejected": -1.5969533920288086, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 3.9507259776993954e-08, | |
| "logits/chosen": -2.0943470001220703, | |
| "logits/rejected": -2.0785202980041504, | |
| "logps/chosen": -405.1844177246094, | |
| "logps/rejected": -463.93841552734375, | |
| "loss": 0.5132, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": -0.9134857058525085, | |
| "rewards/margins": 0.8523229360580444, | |
| "rewards/rejected": -1.7658087015151978, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 3.022313472693447e-08, | |
| "logits/chosen": -2.123349905014038, | |
| "logits/rejected": -2.0609166622161865, | |
| "logps/chosen": -333.0213623046875, | |
| "logps/rejected": -408.7607727050781, | |
| "loss": 0.5302, | |
| "rewards/accuracies": 0.800000011920929, | |
| "rewards/chosen": -0.7784355282783508, | |
| "rewards/margins": 0.8595449328422546, | |
| "rewards/rejected": -1.6379806995391846, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 2.2111614344599684e-08, | |
| "logits/chosen": -2.0923712253570557, | |
| "logits/rejected": -2.0771572589874268, | |
| "logps/chosen": -346.8676452636719, | |
| "logps/rejected": -450.01312255859375, | |
| "loss": 0.5237, | |
| "rewards/accuracies": 0.7562500238418579, | |
| "rewards/chosen": -0.7909084558486938, | |
| "rewards/margins": 0.9126070141792297, | |
| "rewards/rejected": -1.7035152912139893, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 1.521597710086439e-08, | |
| "logits/chosen": -2.203439474105835, | |
| "logits/rejected": -2.1634104251861572, | |
| "logps/chosen": -366.4968566894531, | |
| "logps/rejected": -410.26177978515625, | |
| "loss": 0.5283, | |
| "rewards/accuracies": 0.762499988079071, | |
| "rewards/chosen": -0.7541608810424805, | |
| "rewards/margins": 0.8509491682052612, | |
| "rewards/rejected": -1.6051101684570312, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 9.57301420397924e-09, | |
| "logits/chosen": -2.166147232055664, | |
| "logits/rejected": -2.1166653633117676, | |
| "logps/chosen": -336.30279541015625, | |
| "logps/rejected": -413.140380859375, | |
| "loss": 0.5157, | |
| "rewards/accuracies": 0.71875, | |
| "rewards/chosen": -0.8325687646865845, | |
| "rewards/margins": 0.7907065153121948, | |
| "rewards/rejected": -1.6232753992080688, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 5.212833302556258e-09, | |
| "logits/chosen": -2.073584794998169, | |
| "logits/rejected": -2.025235652923584, | |
| "logps/chosen": -316.84906005859375, | |
| "logps/rejected": -382.339599609375, | |
| "loss": 0.5276, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": -0.7750087976455688, | |
| "rewards/margins": 0.7785266041755676, | |
| "rewards/rejected": -1.5535353422164917, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 2.158697848236607e-09, | |
| "logits/chosen": -2.1052441596984863, | |
| "logits/rejected": -2.039158582687378, | |
| "logps/chosen": -318.8851623535156, | |
| "logps/rejected": -384.3031005859375, | |
| "loss": 0.5063, | |
| "rewards/accuracies": 0.737500011920929, | |
| "rewards/chosen": -0.8155075907707214, | |
| "rewards/margins": 0.7695605158805847, | |
| "rewards/rejected": -1.5850679874420166, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eta": 0.004999999422580004, | |
| "learning_rate": 4.269029751107489e-10, | |
| "logits/chosen": -2.1885509490966797, | |
| "logits/rejected": -2.0868308544158936, | |
| "logps/chosen": -369.0015869140625, | |
| "logps/rejected": -435.66632080078125, | |
| "loss": 0.5031, | |
| "rewards/accuracies": 0.7875000238418579, | |
| "rewards/chosen": -0.7688988447189331, | |
| "rewards/margins": 1.0167477130889893, | |
| "rewards/rejected": -1.7856464385986328, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 478, | |
| "total_flos": 0.0, | |
| "train_loss": 0.5600044772215967, | |
| "train_runtime": 8380.7405, | |
| "train_samples_per_second": 7.295, | |
| "train_steps_per_second": 0.057 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 478, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |