| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9960291197882198, | |
| "eval_steps": 1000, | |
| "global_step": 754, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0026472534745201853, | |
| "grad_norm": 4.30932258831093, | |
| "learning_rate": 6.578947368421052e-09, | |
| "logits/chosen": -2.923454761505127, | |
| "logits/rejected": -3.022336483001709, | |
| "logps/chosen": -491.803955078125, | |
| "logps/rejected": -509.828369140625, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.026472534745201854, | |
| "grad_norm": 4.3524814860184176, | |
| "learning_rate": 6.578947368421052e-08, | |
| "logits/chosen": -2.849837303161621, | |
| "logits/rejected": -2.918842315673828, | |
| "logps/chosen": -482.8021240234375, | |
| "logps/rejected": -468.6262512207031, | |
| "loss": 0.6931, | |
| "rewards/accuracies": 0.4444444477558136, | |
| "rewards/chosen": 1.7793978258850984e-05, | |
| "rewards/margins": 0.00012061676534358412, | |
| "rewards/rejected": -0.00010282275616191328, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05294506949040371, | |
| "grad_norm": 4.058367941679181, | |
| "learning_rate": 1.3157894736842104e-07, | |
| "logits/chosen": -2.8850979804992676, | |
| "logits/rejected": -2.9686431884765625, | |
| "logps/chosen": -492.1334533691406, | |
| "logps/rejected": -473.5765075683594, | |
| "loss": 0.6929, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": 0.00013926837709732354, | |
| "rewards/margins": 0.0005674505373463035, | |
| "rewards/rejected": -0.0004281821602489799, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07941760423560557, | |
| "grad_norm": 4.100822256989, | |
| "learning_rate": 1.9736842105263157e-07, | |
| "logits/chosen": -2.868659496307373, | |
| "logits/rejected": -2.9573891162872314, | |
| "logps/chosen": -475.2897033691406, | |
| "logps/rejected": -473.85064697265625, | |
| "loss": 0.6906, | |
| "rewards/accuracies": 0.715624988079071, | |
| "rewards/chosen": 0.003203944070264697, | |
| "rewards/margins": 0.006764715071767569, | |
| "rewards/rejected": -0.0035607716999948025, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.10589013898080742, | |
| "grad_norm": 4.224597004906648, | |
| "learning_rate": 2.631578947368421e-07, | |
| "logits/chosen": -2.880610704421997, | |
| "logits/rejected": -2.9531335830688477, | |
| "logps/chosen": -477.7626953125, | |
| "logps/rejected": -471.15203857421875, | |
| "loss": 0.6838, | |
| "rewards/accuracies": 0.7906249761581421, | |
| "rewards/chosen": 0.0102998660877347, | |
| "rewards/margins": 0.02004994824528694, | |
| "rewards/rejected": -0.009750082157552242, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.13236267372600927, | |
| "grad_norm": 4.481790796785453, | |
| "learning_rate": 3.2894736842105264e-07, | |
| "logits/chosen": -2.910224199295044, | |
| "logits/rejected": -2.966555118560791, | |
| "logps/chosen": -496.03704833984375, | |
| "logps/rejected": -483.69970703125, | |
| "loss": 0.6705, | |
| "rewards/accuracies": 0.8031250238418579, | |
| "rewards/chosen": 0.02056797966361046, | |
| "rewards/margins": 0.04747764393687248, | |
| "rewards/rejected": -0.026909660547971725, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15883520847121113, | |
| "grad_norm": 4.727420695183077, | |
| "learning_rate": 3.9473684210526315e-07, | |
| "logits/chosen": -2.908859968185425, | |
| "logits/rejected": -2.9846339225769043, | |
| "logps/chosen": -487.971923828125, | |
| "logps/rejected": -487.698486328125, | |
| "loss": 0.6321, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": 0.047008998692035675, | |
| "rewards/margins": 0.12849071621894836, | |
| "rewards/rejected": -0.08148171752691269, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.18530774321641297, | |
| "grad_norm": 5.3651750883138645, | |
| "learning_rate": 4.6052631578947365e-07, | |
| "logits/chosen": -2.9276702404022217, | |
| "logits/rejected": -2.958789348602295, | |
| "logps/chosen": -488.4877014160156, | |
| "logps/rejected": -489.53546142578125, | |
| "loss": 0.5937, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": 0.025530118495225906, | |
| "rewards/margins": 0.22059020400047302, | |
| "rewards/rejected": -0.19506008923053741, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.21178027796161483, | |
| "grad_norm": 3.6311630185736297, | |
| "learning_rate": 4.999570604073014e-07, | |
| "logits/chosen": -2.923982620239258, | |
| "logits/rejected": -2.9833145141601562, | |
| "logps/chosen": -503.56134033203125, | |
| "logps/rejected": -552.4190673828125, | |
| "loss": 0.4934, | |
| "rewards/accuracies": 0.840624988079071, | |
| "rewards/chosen": -0.13993698358535767, | |
| "rewards/margins": 0.5944468975067139, | |
| "rewards/rejected": -0.7343839406967163, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.23825281270681667, | |
| "grad_norm": 4.662865437270245, | |
| "learning_rate": 4.994741593733563e-07, | |
| "logits/chosen": -2.9195408821105957, | |
| "logits/rejected": -2.958824872970581, | |
| "logps/chosen": -534.3409423828125, | |
| "logps/rejected": -607.5826416015625, | |
| "loss": 0.4331, | |
| "rewards/accuracies": 0.871874988079071, | |
| "rewards/chosen": -0.41773176193237305, | |
| "rewards/margins": 0.9263063669204712, | |
| "rewards/rejected": -1.3440382480621338, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.26472534745201853, | |
| "grad_norm": 4.763826147961603, | |
| "learning_rate": 4.984557228946769e-07, | |
| "logits/chosen": -2.815560817718506, | |
| "logits/rejected": -2.8625988960266113, | |
| "logps/chosen": -611.6197509765625, | |
| "logps/rejected": -758.5867309570312, | |
| "loss": 0.3704, | |
| "rewards/accuracies": 0.8343750238418579, | |
| "rewards/chosen": -1.2842621803283691, | |
| "rewards/margins": 1.4864912033081055, | |
| "rewards/rejected": -2.7707533836364746, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.29119788219722037, | |
| "grad_norm": 4.297222720122574, | |
| "learning_rate": 4.969039372050355e-07, | |
| "logits/chosen": -2.787930488586426, | |
| "logits/rejected": -2.8286757469177246, | |
| "logps/chosen": -622.7388305664062, | |
| "logps/rejected": -840.2802734375, | |
| "loss": 0.3234, | |
| "rewards/accuracies": 0.8531249761581421, | |
| "rewards/chosen": -1.4110974073410034, | |
| "rewards/margins": 2.2375450134277344, | |
| "rewards/rejected": -3.648642063140869, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.31767041694242226, | |
| "grad_norm": 3.7535819941587065, | |
| "learning_rate": 4.948221334560093e-07, | |
| "logits/chosen": -2.831071615219116, | |
| "logits/rejected": -2.843599796295166, | |
| "logps/chosen": -592.7720947265625, | |
| "logps/rejected": -807.1361083984375, | |
| "loss": 0.3176, | |
| "rewards/accuracies": 0.8531249761581421, | |
| "rewards/chosen": -1.023418664932251, | |
| "rewards/margins": 2.2339682579040527, | |
| "rewards/rejected": -3.2573866844177246, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3441429516876241, | |
| "grad_norm": 4.4946406733821185, | |
| "learning_rate": 4.922147805661402e-07, | |
| "logits/chosen": -2.83577823638916, | |
| "logits/rejected": -2.875858783721924, | |
| "logps/chosen": -629.0977783203125, | |
| "logps/rejected": -913.8206787109375, | |
| "loss": 0.2875, | |
| "rewards/accuracies": 0.862500011920929, | |
| "rewards/chosen": -1.3898423910140991, | |
| "rewards/margins": 2.828009605407715, | |
| "rewards/rejected": -4.2178521156311035, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.37061548643282594, | |
| "grad_norm": 4.541566634311875, | |
| "learning_rate": 4.890874756276999e-07, | |
| "logits/chosen": -2.7425622940063477, | |
| "logits/rejected": -2.8039002418518066, | |
| "logps/chosen": -634.9403686523438, | |
| "logps/rejected": -911.8308715820312, | |
| "loss": 0.2831, | |
| "rewards/accuracies": 0.8500000238418579, | |
| "rewards/chosen": -1.5872424840927124, | |
| "rewards/margins": 2.6655924320220947, | |
| "rewards/rejected": -4.252835273742676, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3970880211780278, | |
| "grad_norm": 3.3580041177607978, | |
| "learning_rate": 4.854469318916532e-07, | |
| "logits/chosen": -2.8008108139038086, | |
| "logits/rejected": -2.8558998107910156, | |
| "logps/chosen": -646.1107177734375, | |
| "logps/rejected": -980.4015502929688, | |
| "loss": 0.2668, | |
| "rewards/accuracies": 0.9156249761581421, | |
| "rewards/chosen": -1.5013834238052368, | |
| "rewards/margins": 3.364828109741211, | |
| "rewards/rejected": -4.866211414337158, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.42356055592322966, | |
| "grad_norm": 4.675914141471813, | |
| "learning_rate": 4.8130096435661e-07, | |
| "logits/chosen": -2.80204176902771, | |
| "logits/rejected": -2.8504693508148193, | |
| "logps/chosen": -631.5078125, | |
| "logps/rejected": -925.1160888671875, | |
| "loss": 0.2519, | |
| "rewards/accuracies": 0.893750011920929, | |
| "rewards/chosen": -1.4621139764785767, | |
| "rewards/margins": 2.9786953926086426, | |
| "rewards/rejected": -4.440809726715088, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4500330906684315, | |
| "grad_norm": 6.483714450702139, | |
| "learning_rate": 4.766584729927049e-07, | |
| "logits/chosen": -2.768064498901367, | |
| "logits/rejected": -2.8086233139038086, | |
| "logps/chosen": -676.531982421875, | |
| "logps/rejected": -1029.0057373046875, | |
| "loss": 0.2799, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -1.8374897241592407, | |
| "rewards/margins": 3.5798168182373047, | |
| "rewards/rejected": -5.417306900024414, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.47650562541363334, | |
| "grad_norm": 5.396515481877322, | |
| "learning_rate": 4.7152942363641345e-07, | |
| "logits/chosen": -2.7385268211364746, | |
| "logits/rejected": -2.760967493057251, | |
| "logps/chosen": -642.5712890625, | |
| "logps/rejected": -997.8201904296875, | |
| "loss": 0.2558, | |
| "rewards/accuracies": 0.8968750238418579, | |
| "rewards/chosen": -1.627642273902893, | |
| "rewards/margins": 3.5839648246765137, | |
| "rewards/rejected": -5.211607456207275, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.5029781601588352, | |
| "grad_norm": 5.261191188089043, | |
| "learning_rate": 4.6592482659732045e-07, | |
| "logits/chosen": -2.7847423553466797, | |
| "logits/rejected": -2.8355870246887207, | |
| "logps/chosen": -674.9547729492188, | |
| "logps/rejected": -972.1364135742188, | |
| "loss": 0.2536, | |
| "rewards/accuracies": 0.8968750238418579, | |
| "rewards/chosen": -1.8109451532363892, | |
| "rewards/margins": 3.033442258834839, | |
| "rewards/rejected": -4.844388008117676, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.5294506949040371, | |
| "grad_norm": 4.9133890145951495, | |
| "learning_rate": 4.5985671302276166e-07, | |
| "logits/chosen": -2.7557764053344727, | |
| "logits/rejected": -2.8086917400360107, | |
| "logps/chosen": -681.2697143554688, | |
| "logps/rejected": -989.4937744140625, | |
| "loss": 0.2602, | |
| "rewards/accuracies": 0.890625, | |
| "rewards/chosen": -1.8896667957305908, | |
| "rewards/margins": 3.2132556438446045, | |
| "rewards/rejected": -5.102922439575195, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5559232296492389, | |
| "grad_norm": 4.892431641268176, | |
| "learning_rate": 4.533381090710776e-07, | |
| "logits/chosen": -2.7014384269714355, | |
| "logits/rejected": -2.7446448802948, | |
| "logps/chosen": -670.3836669921875, | |
| "logps/rejected": -988.2939453125, | |
| "loss": 0.2323, | |
| "rewards/accuracies": 0.909375011920929, | |
| "rewards/chosen": -1.8962217569351196, | |
| "rewards/margins": 3.261091947555542, | |
| "rewards/rejected": -5.157313823699951, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.5823957643944407, | |
| "grad_norm": 4.5229078992046166, | |
| "learning_rate": 4.463830079489196e-07, | |
| "logits/chosen": -2.7127437591552734, | |
| "logits/rejected": -2.735060930252075, | |
| "logps/chosen": -697.1866455078125, | |
| "logps/rejected": -1054.2486572265625, | |
| "loss": 0.2292, | |
| "rewards/accuracies": 0.9156249761581421, | |
| "rewards/chosen": -2.145174026489258, | |
| "rewards/margins": 3.7120985984802246, | |
| "rewards/rejected": -5.857272148132324, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6088682991396426, | |
| "grad_norm": 5.024605464390731, | |
| "learning_rate": 4.390063398726356e-07, | |
| "logits/chosen": -2.6947622299194336, | |
| "logits/rejected": -2.6960158348083496, | |
| "logps/chosen": -719.8890380859375, | |
| "logps/rejected": -1086.1702880859375, | |
| "loss": 0.2353, | |
| "rewards/accuracies": 0.921875, | |
| "rewards/chosen": -2.32169771194458, | |
| "rewards/margins": 3.803462266921997, | |
| "rewards/rejected": -6.12515926361084, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.6353408338848445, | |
| "grad_norm": 4.497134908678504, | |
| "learning_rate": 4.3122394001821657e-07, | |
| "logits/chosen": -2.699171781539917, | |
| "logits/rejected": -2.7273011207580566, | |
| "logps/chosen": -715.7581176757812, | |
| "logps/rejected": -1008.9625244140625, | |
| "loss": 0.2481, | |
| "rewards/accuracies": 0.875, | |
| "rewards/chosen": -2.174252986907959, | |
| "rewards/margins": 3.0458567142486572, | |
| "rewards/rejected": -5.220109462738037, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6618133686300464, | |
| "grad_norm": 4.8761682045650625, | |
| "learning_rate": 4.2305251452860566e-07, | |
| "logits/chosen": -2.7068121433258057, | |
| "logits/rejected": -2.7410686016082764, | |
| "logps/chosen": -667.6207885742188, | |
| "logps/rejected": -1025.73828125, | |
| "loss": 0.2279, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": -1.91938054561615, | |
| "rewards/margins": 3.5805881023406982, | |
| "rewards/rejected": -5.4999680519104, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.6882859033752482, | |
| "grad_norm": 6.933116753661563, | |
| "learning_rate": 4.1450960465134024e-07, | |
| "logits/chosen": -2.7003915309906006, | |
| "logits/rejected": -2.7421202659606934, | |
| "logps/chosen": -725.2587890625, | |
| "logps/rejected": -1111.21533203125, | |
| "loss": 0.2322, | |
| "rewards/accuracies": 0.90625, | |
| "rewards/chosen": -2.4127094745635986, | |
| "rewards/margins": 3.868704319000244, | |
| "rewards/rejected": -6.28141450881958, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.71475843812045, | |
| "grad_norm": 5.180639131738146, | |
| "learning_rate": 4.0561354908350977e-07, | |
| "logits/chosen": -2.7455577850341797, | |
| "logits/rejected": -2.768336057662964, | |
| "logps/chosen": -686.0287475585938, | |
| "logps/rejected": -1041.7474365234375, | |
| "loss": 0.2406, | |
| "rewards/accuracies": 0.925000011920929, | |
| "rewards/chosen": -2.0675699710845947, | |
| "rewards/margins": 3.5837669372558594, | |
| "rewards/rejected": -5.651337146759033, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.7412309728656519, | |
| "grad_norm": 6.760828116437296, | |
| "learning_rate": 3.963834446048644e-07, | |
| "logits/chosen": -2.7979235649108887, | |
| "logits/rejected": -2.795571804046631, | |
| "logps/chosen": -757.2296142578125, | |
| "logps/rejected": -1096.9935302734375, | |
| "loss": 0.235, | |
| "rewards/accuracies": 0.878125011920929, | |
| "rewards/chosen": -2.414430618286133, | |
| "rewards/margins": 3.682656764984131, | |
| "rewards/rejected": -6.097087383270264, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7677035076108537, | |
| "grad_norm": 5.924062579551431, | |
| "learning_rate": 3.868391050835793e-07, | |
| "logits/chosen": -2.707353353500366, | |
| "logits/rejected": -2.708177089691162, | |
| "logps/chosen": -735.2392578125, | |
| "logps/rejected": -1096.3634033203125, | |
| "loss": 0.2165, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -2.4990901947021484, | |
| "rewards/margins": 3.737868547439575, | |
| "rewards/rejected": -6.2369585037231445, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.7941760423560555, | |
| "grad_norm": 6.316712123280126, | |
| "learning_rate": 3.770010189426761e-07, | |
| "logits/chosen": -2.696866512298584, | |
| "logits/rejected": -2.720412015914917, | |
| "logps/chosen": -788.0701904296875, | |
| "logps/rejected": -1174.900390625, | |
| "loss": 0.2278, | |
| "rewards/accuracies": 0.893750011920929, | |
| "rewards/chosen": -2.9900288581848145, | |
| "rewards/margins": 3.7810866832733154, | |
| "rewards/rejected": -6.771115779876709, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.8206485771012575, | |
| "grad_norm": 5.395016037418297, | |
| "learning_rate": 3.66890305178407e-07, | |
| "logits/chosen": -2.716305732727051, | |
| "logits/rejected": -2.716601848602295, | |
| "logps/chosen": -742.38232421875, | |
| "logps/rejected": -1122.4625244140625, | |
| "loss": 0.2287, | |
| "rewards/accuracies": 0.90625, | |
| "rewards/chosen": -2.601231098175049, | |
| "rewards/margins": 3.891951084136963, | |
| "rewards/rejected": -6.4931817054748535, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.8471211118464593, | |
| "grad_norm": 6.246149943149367, | |
| "learning_rate": 3.565286680250138e-07, | |
| "logits/chosen": -2.748223066329956, | |
| "logits/rejected": -2.7793920040130615, | |
| "logps/chosen": -724.46484375, | |
| "logps/rejected": -1080.5015869140625, | |
| "loss": 0.2089, | |
| "rewards/accuracies": 0.9156249761581421, | |
| "rewards/chosen": -2.3395774364471436, | |
| "rewards/margins": 3.757481813430786, | |
| "rewards/rejected": -6.097059726715088, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8735936465916612, | |
| "grad_norm": 6.876926943284203, | |
| "learning_rate": 3.4593835036318225e-07, | |
| "logits/chosen": -2.691286563873291, | |
| "logits/rejected": -2.7156758308410645, | |
| "logps/chosen": -759.7913208007812, | |
| "logps/rejected": -1132.6741943359375, | |
| "loss": 0.2212, | |
| "rewards/accuracies": 0.921875, | |
| "rewards/chosen": -2.764772415161133, | |
| "rewards/margins": 3.8325703144073486, | |
| "rewards/rejected": -6.597343444824219, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.900066181336863, | |
| "grad_norm": 7.266749393582284, | |
| "learning_rate": 3.35142085972207e-07, | |
| "logits/chosen": -2.6851227283477783, | |
| "logits/rejected": -2.722938060760498, | |
| "logps/chosen": -754.1365356445312, | |
| "logps/rejected": -1132.1871337890625, | |
| "loss": 0.218, | |
| "rewards/accuracies": 0.8843749761581421, | |
| "rewards/chosen": -2.7608418464660645, | |
| "rewards/margins": 3.8383007049560547, | |
| "rewards/rejected": -6.599142551422119, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.9265387160820648, | |
| "grad_norm": 5.8828162546054426, | |
| "learning_rate": 3.2416305072836555e-07, | |
| "logits/chosen": -2.703392267227173, | |
| "logits/rejected": -2.717411756515503, | |
| "logps/chosen": -759.862060546875, | |
| "logps/rejected": -1137.0098876953125, | |
| "loss": 0.2033, | |
| "rewards/accuracies": 0.909375011920929, | |
| "rewards/chosen": -2.6198081970214844, | |
| "rewards/margins": 3.8972651958465576, | |
| "rewards/rejected": -6.517073154449463, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9530112508272667, | |
| "grad_norm": 6.02673405904732, | |
| "learning_rate": 3.1302481285426197e-07, | |
| "logits/chosen": -2.7140469551086426, | |
| "logits/rejected": -2.7299208641052246, | |
| "logps/chosen": -727.9823608398438, | |
| "logps/rejected": -1152.2041015625, | |
| "loss": 0.2146, | |
| "rewards/accuracies": 0.921875, | |
| "rewards/chosen": -2.4466357231140137, | |
| "rewards/margins": 4.253617763519287, | |
| "rewards/rejected": -6.700253486633301, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9794837855724685, | |
| "grad_norm": 5.891168724337083, | |
| "learning_rate": 3.017512823259373e-07, | |
| "logits/chosen": -2.7166543006896973, | |
| "logits/rejected": -2.7297706604003906, | |
| "logps/chosen": -730.9952392578125, | |
| "logps/rejected": -1117.7025146484375, | |
| "loss": 0.2011, | |
| "rewards/accuracies": 0.9281250238418579, | |
| "rewards/chosen": -2.5781686305999756, | |
| "rewards/margins": 4.00323486328125, | |
| "rewards/rejected": -6.5814032554626465, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.0059563203176705, | |
| "grad_norm": 5.4096108526897115, | |
| "learning_rate": 2.9036665954635264e-07, | |
| "logits/chosen": -2.7353827953338623, | |
| "logits/rejected": -2.765237331390381, | |
| "logps/chosen": -769.5081176757812, | |
| "logps/rejected": -1209.970947265625, | |
| "loss": 0.2135, | |
| "rewards/accuracies": 0.940625011920929, | |
| "rewards/chosen": -2.7778820991516113, | |
| "rewards/margins": 4.453129291534424, | |
| "rewards/rejected": -7.231011390686035, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.0324288550628722, | |
| "grad_norm": 6.969996057859779, | |
| "learning_rate": 2.7889538339542523e-07, | |
| "logits/chosen": -2.715958833694458, | |
| "logits/rejected": -2.741973638534546, | |
| "logps/chosen": -760.3748779296875, | |
| "logps/rejected": -1181.979248046875, | |
| "loss": 0.1997, | |
| "rewards/accuracies": 0.940625011920929, | |
| "rewards/chosen": -2.8078932762145996, | |
| "rewards/margins": 4.318324089050293, | |
| "rewards/rejected": -7.126217842102051, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.0589013898080741, | |
| "grad_norm": 6.27267698414733, | |
| "learning_rate": 2.6736207876813643e-07, | |
| "logits/chosen": -2.728332042694092, | |
| "logits/rejected": -2.7608368396759033, | |
| "logps/chosen": -768.7643432617188, | |
| "logps/rejected": -1208.543701171875, | |
| "loss": 0.1963, | |
| "rewards/accuracies": 0.909375011920929, | |
| "rewards/chosen": -2.7519562244415283, | |
| "rewards/margins": 4.398838996887207, | |
| "rewards/rejected": -7.150795936584473, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0853739245532759, | |
| "grad_norm": 6.331116230324871, | |
| "learning_rate": 2.5579150371332953e-07, | |
| "logits/chosen": -2.7396435737609863, | |
| "logits/rejected": -2.7421391010284424, | |
| "logps/chosen": -809.5894775390625, | |
| "logps/rejected": -1248.327880859375, | |
| "loss": 0.1821, | |
| "rewards/accuracies": 0.940625011920929, | |
| "rewards/chosen": -3.0432112216949463, | |
| "rewards/margins": 4.567535400390625, | |
| "rewards/rejected": -7.61074686050415, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.1118464592984778, | |
| "grad_norm": 6.507894575900424, | |
| "learning_rate": 2.4420849628667045e-07, | |
| "logits/chosen": -2.7221198081970215, | |
| "logits/rejected": -2.7356066703796387, | |
| "logps/chosen": -795.4509887695312, | |
| "logps/rejected": -1229.848876953125, | |
| "loss": 0.205, | |
| "rewards/accuracies": 0.940625011920929, | |
| "rewards/chosen": -3.027529716491699, | |
| "rewards/margins": 4.336198806762695, | |
| "rewards/rejected": -7.3637285232543945, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.1383189940436798, | |
| "grad_norm": 6.866126936883296, | |
| "learning_rate": 2.3263792123186352e-07, | |
| "logits/chosen": -2.7347493171691895, | |
| "logits/rejected": -2.7523412704467773, | |
| "logps/chosen": -788.9684448242188, | |
| "logps/rejected": -1187.0565185546875, | |
| "loss": 0.2132, | |
| "rewards/accuracies": 0.8843749761581421, | |
| "rewards/chosen": -2.9675726890563965, | |
| "rewards/margins": 4.024303913116455, | |
| "rewards/rejected": -6.99187707901001, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.1647915287888815, | |
| "grad_norm": 5.664097996574406, | |
| "learning_rate": 2.211046166045748e-07, | |
| "logits/chosen": -2.699467420578003, | |
| "logits/rejected": -2.715172529220581, | |
| "logps/chosen": -787.1725463867188, | |
| "logps/rejected": -1219.5947265625, | |
| "loss": 0.1992, | |
| "rewards/accuracies": 0.909375011920929, | |
| "rewards/chosen": -2.9378509521484375, | |
| "rewards/margins": 4.4215216636657715, | |
| "rewards/rejected": -7.359372138977051, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.1912640635340834, | |
| "grad_norm": 5.8438221376315, | |
| "learning_rate": 2.096333404536474e-07, | |
| "logits/chosen": -2.6697657108306885, | |
| "logits/rejected": -2.6788392066955566, | |
| "logps/chosen": -816.9085693359375, | |
| "logps/rejected": -1204.7374267578125, | |
| "loss": 0.1939, | |
| "rewards/accuracies": 0.925000011920929, | |
| "rewards/chosen": -3.2424380779266357, | |
| "rewards/margins": 4.097726821899414, | |
| "rewards/rejected": -7.340165138244629, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.2177365982792852, | |
| "grad_norm": 6.534950899522826, | |
| "learning_rate": 1.982487176740627e-07, | |
| "logits/chosen": -2.6479620933532715, | |
| "logits/rejected": -2.686981201171875, | |
| "logps/chosen": -760.5189208984375, | |
| "logps/rejected": -1172.6590576171875, | |
| "loss": 0.2034, | |
| "rewards/accuracies": 0.9281250238418579, | |
| "rewards/chosen": -2.8477563858032227, | |
| "rewards/margins": 4.02510929107666, | |
| "rewards/rejected": -6.872865200042725, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.244209133024487, | |
| "grad_norm": 5.919455336024619, | |
| "learning_rate": 1.8697518714573804e-07, | |
| "logits/chosen": -2.7183327674865723, | |
| "logits/rejected": -2.7170839309692383, | |
| "logps/chosen": -790.9767456054688, | |
| "logps/rejected": -1145.7747802734375, | |
| "loss": 0.1854, | |
| "rewards/accuracies": 0.887499988079071, | |
| "rewards/chosen": -2.942429304122925, | |
| "rewards/margins": 3.9219982624053955, | |
| "rewards/rejected": -6.864427089691162, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.270681667769689, | |
| "grad_norm": 8.510303191565324, | |
| "learning_rate": 1.758369492716345e-07, | |
| "logits/chosen": -2.7107906341552734, | |
| "logits/rejected": -2.7333004474639893, | |
| "logps/chosen": -831.3541870117188, | |
| "logps/rejected": -1250.9085693359375, | |
| "loss": 0.204, | |
| "rewards/accuracies": 0.9156249761581421, | |
| "rewards/chosen": -3.3519797325134277, | |
| "rewards/margins": 4.182629585266113, | |
| "rewards/rejected": -7.534609317779541, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.2971542025148908, | |
| "grad_norm": 6.158432969646956, | |
| "learning_rate": 1.648579140277931e-07, | |
| "logits/chosen": -2.7101292610168457, | |
| "logits/rejected": -2.715567111968994, | |
| "logps/chosen": -815.2239990234375, | |
| "logps/rejected": -1204.00830078125, | |
| "loss": 0.2016, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": -3.184443712234497, | |
| "rewards/margins": 4.096536159515381, | |
| "rewards/rejected": -7.280980110168457, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.3236267372600927, | |
| "grad_norm": 6.688172890404941, | |
| "learning_rate": 1.5406164963681773e-07, | |
| "logits/chosen": -2.6899495124816895, | |
| "logits/rejected": -2.697331666946411, | |
| "logps/chosen": -788.6990356445312, | |
| "logps/rejected": -1184.682861328125, | |
| "loss": 0.2028, | |
| "rewards/accuracies": 0.918749988079071, | |
| "rewards/chosen": -2.907625436782837, | |
| "rewards/margins": 4.130288124084473, | |
| "rewards/rejected": -7.037914276123047, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.3500992720052944, | |
| "grad_norm": 6.177665969794554, | |
| "learning_rate": 1.4347133197498618e-07, | |
| "logits/chosen": -2.720623016357422, | |
| "logits/rejected": -2.713228702545166, | |
| "logps/chosen": -813.38427734375, | |
| "logps/rejected": -1158.912353515625, | |
| "loss": 0.1947, | |
| "rewards/accuracies": 0.8812500238418579, | |
| "rewards/chosen": -3.203073024749756, | |
| "rewards/margins": 3.7714266777038574, | |
| "rewards/rejected": -6.974499702453613, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.3765718067504964, | |
| "grad_norm": 7.6140239843164785, | |
| "learning_rate": 1.3310969482159296e-07, | |
| "logits/chosen": -2.7323203086853027, | |
| "logits/rejected": -2.720503330230713, | |
| "logps/chosen": -822.5255737304688, | |
| "logps/rejected": -1203.3837890625, | |
| "loss": 0.206, | |
| "rewards/accuracies": 0.8843749761581421, | |
| "rewards/chosen": -3.2711524963378906, | |
| "rewards/margins": 4.040999889373779, | |
| "rewards/rejected": -7.312152862548828, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.4030443414956983, | |
| "grad_norm": 7.395646356776199, | |
| "learning_rate": 1.2299898105732384e-07, | |
| "logits/chosen": -2.664844036102295, | |
| "logits/rejected": -2.6827051639556885, | |
| "logps/chosen": -797.8216552734375, | |
| "logps/rejected": -1175.84326171875, | |
| "loss": 0.1975, | |
| "rewards/accuracies": 0.90625, | |
| "rewards/chosen": -3.197530746459961, | |
| "rewards/margins": 3.875856876373291, | |
| "rewards/rejected": -7.07338809967041, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.4295168762409, | |
| "grad_norm": 5.37023800702416, | |
| "learning_rate": 1.1316089491642075e-07, | |
| "logits/chosen": -2.6801838874816895, | |
| "logits/rejected": -2.7150120735168457, | |
| "logps/chosen": -782.3211669921875, | |
| "logps/rejected": -1218.0201416015625, | |
| "loss": 0.177, | |
| "rewards/accuracies": 0.953125, | |
| "rewards/chosen": -3.0446789264678955, | |
| "rewards/margins": 4.468942642211914, | |
| "rewards/rejected": -7.5136213302612305, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.4559894109861018, | |
| "grad_norm": 6.848159521074984, | |
| "learning_rate": 1.0361655539513564e-07, | |
| "logits/chosen": -2.7019715309143066, | |
| "logits/rejected": -2.717634677886963, | |
| "logps/chosen": -816.9815063476562, | |
| "logps/rejected": -1239.1295166015625, | |
| "loss": 0.1866, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": -3.2783799171447754, | |
| "rewards/margins": 4.237971305847168, | |
| "rewards/rejected": -7.516351222991943, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.4824619457313037, | |
| "grad_norm": 6.1903290336235965, | |
| "learning_rate": 9.438645091649028e-08, | |
| "logits/chosen": -2.703291416168213, | |
| "logits/rejected": -2.717991590499878, | |
| "logps/chosen": -835.4637451171875, | |
| "logps/rejected": -1262.481201171875, | |
| "loss": 0.1941, | |
| "rewards/accuracies": 0.909375011920929, | |
| "rewards/chosen": -3.5229759216308594, | |
| "rewards/margins": 4.230559349060059, | |
| "rewards/rejected": -7.753535270690918, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.5089344804765057, | |
| "grad_norm": 7.319396362423722, | |
| "learning_rate": 8.549039534865979e-08, | |
| "logits/chosen": -2.6856772899627686, | |
| "logits/rejected": -2.7186999320983887, | |
| "logps/chosen": -839.0086059570312, | |
| "logps/rejected": -1260.779541015625, | |
| "loss": 0.1939, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": -3.4230124950408936, | |
| "rewards/margins": 4.303016185760498, | |
| "rewards/rejected": -7.7260284423828125, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.5354070152217076, | |
| "grad_norm": 6.830716936510353, | |
| "learning_rate": 7.694748547139429e-08, | |
| "logits/chosen": -2.690796136856079, | |
| "logits/rejected": -2.6854541301727295, | |
| "logps/chosen": -858.9547119140625, | |
| "logps/rejected": -1268.125732421875, | |
| "loss": 0.1822, | |
| "rewards/accuracies": 0.925000011920929, | |
| "rewards/chosen": -3.606029510498047, | |
| "rewards/margins": 4.35842752456665, | |
| "rewards/rejected": -7.964457035064697, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.5618795499669094, | |
| "grad_norm": 7.344475965596424, | |
| "learning_rate": 6.877605998178343e-08, | |
| "logits/chosen": -2.6852376461029053, | |
| "logits/rejected": -2.6851906776428223, | |
| "logps/chosen": -825.2852783203125, | |
| "logps/rejected": -1217.6024169921875, | |
| "loss": 0.1937, | |
| "rewards/accuracies": 0.9312499761581421, | |
| "rewards/chosen": -3.3112761974334717, | |
| "rewards/margins": 4.0954132080078125, | |
| "rewards/rejected": -7.4066901206970215, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.588352084712111, | |
| "grad_norm": 6.774677658543177, | |
| "learning_rate": 6.099366012736437e-08, | |
| "logits/chosen": -2.6621475219726562, | |
| "logits/rejected": -2.6802525520324707, | |
| "logps/chosen": -823.2789306640625, | |
| "logps/rejected": -1230.999755859375, | |
| "loss": 0.1879, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -3.3858489990234375, | |
| "rewards/margins": 4.153179168701172, | |
| "rewards/rejected": -7.539028167724609, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.614824619457313, | |
| "grad_norm": 6.63238227242087, | |
| "learning_rate": 5.3616992051080415e-08, | |
| "logits/chosen": -2.677306890487671, | |
| "logits/rejected": -2.712442636489868, | |
| "logps/chosen": -788.0901489257812, | |
| "logps/rejected": -1236.086181640625, | |
| "loss": 0.1818, | |
| "rewards/accuracies": 0.925000011920929, | |
| "rewards/chosen": -3.1388726234436035, | |
| "rewards/margins": 4.399405002593994, | |
| "rewards/rejected": -7.538276672363281, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.641297154202515, | |
| "grad_norm": 7.377420865927689, | |
| "learning_rate": 4.666189092892245e-08, | |
| "logits/chosen": -2.681398630142212, | |
| "logits/rejected": -2.7074177265167236, | |
| "logps/chosen": -808.8873291015625, | |
| "logps/rejected": -1222.2945556640625, | |
| "loss": 0.192, | |
| "rewards/accuracies": 0.90625, | |
| "rewards/chosen": -3.247420072555542, | |
| "rewards/margins": 4.086552619934082, | |
| "rewards/rejected": -7.3339738845825195, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.6677696889477167, | |
| "grad_norm": 8.778568268489295, | |
| "learning_rate": 4.0143286977238345e-08, | |
| "logits/chosen": -2.732050895690918, | |
| "logits/rejected": -2.743983507156372, | |
| "logps/chosen": -833.8406372070312, | |
| "logps/rejected": -1255.4339599609375, | |
| "loss": 0.2052, | |
| "rewards/accuracies": 0.909375011920929, | |
| "rewards/chosen": -3.412902355194092, | |
| "rewards/margins": 4.319065570831299, | |
| "rewards/rejected": -7.731966972351074, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.6942422236929184, | |
| "grad_norm": 9.757233734434815, | |
| "learning_rate": 3.407517340267957e-08, | |
| "logits/chosen": -2.704099655151367, | |
| "logits/rejected": -2.747129440307617, | |
| "logps/chosen": -825.1597900390625, | |
| "logps/rejected": -1279.721923828125, | |
| "loss": 0.1969, | |
| "rewards/accuracies": 0.921875, | |
| "rewards/chosen": -3.3202297687530518, | |
| "rewards/margins": 4.6456804275512695, | |
| "rewards/rejected": -7.965909481048584, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.7207147584381204, | |
| "grad_norm": 5.790336419749713, | |
| "learning_rate": 2.847057636358663e-08, | |
| "logits/chosen": -2.666538715362549, | |
| "logits/rejected": -2.6833436489105225, | |
| "logps/chosen": -791.0831298828125, | |
| "logps/rejected": -1239.421875, | |
| "loss": 0.2025, | |
| "rewards/accuracies": 0.8999999761581421, | |
| "rewards/chosen": -3.060492753982544, | |
| "rewards/margins": 4.618899345397949, | |
| "rewards/rejected": -7.679392337799072, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.7471872931833223, | |
| "grad_norm": 7.090658258583613, | |
| "learning_rate": 2.3341527007295107e-08, | |
| "logits/chosen": -2.724691867828369, | |
| "logits/rejected": -2.7381081581115723, | |
| "logps/chosen": -795.6064453125, | |
| "logps/rejected": -1221.3892822265625, | |
| "loss": 0.1934, | |
| "rewards/accuracies": 0.918749988079071, | |
| "rewards/chosen": -3.000559091567993, | |
| "rewards/margins": 4.30582332611084, | |
| "rewards/rejected": -7.306382656097412, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.7736598279285243, | |
| "grad_norm": 7.043130947329383, | |
| "learning_rate": 1.8699035643389927e-08, | |
| "logits/chosen": -2.6909706592559814, | |
| "logits/rejected": -2.706798791885376, | |
| "logps/chosen": -813.7172241210938, | |
| "logps/rejected": -1210.0374755859375, | |
| "loss": 0.1935, | |
| "rewards/accuracies": 0.9375, | |
| "rewards/chosen": -3.150881290435791, | |
| "rewards/margins": 4.167717933654785, | |
| "rewards/rejected": -7.318598747253418, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.800132362673726, | |
| "grad_norm": 8.93025020392657, | |
| "learning_rate": 1.4553068108346778e-08, | |
| "logits/chosen": -2.7232556343078613, | |
| "logits/rejected": -2.758570671081543, | |
| "logps/chosen": -816.0042724609375, | |
| "logps/rejected": -1228.6907958984375, | |
| "loss": 0.1969, | |
| "rewards/accuracies": 0.9156249761581421, | |
| "rewards/chosen": -3.222668409347534, | |
| "rewards/margins": 4.130661487579346, | |
| "rewards/rejected": -7.353329658508301, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.8266048974189277, | |
| "grad_norm": 8.105462657724475, | |
| "learning_rate": 1.0912524372300031e-08, | |
| "logits/chosen": -2.6958508491516113, | |
| "logits/rejected": -2.7145791053771973, | |
| "logps/chosen": -800.2642211914062, | |
| "logps/rejected": -1204.597412109375, | |
| "loss": 0.1979, | |
| "rewards/accuracies": 0.9281250238418579, | |
| "rewards/chosen": -3.0551960468292236, | |
| "rewards/margins": 4.35771369934082, | |
| "rewards/rejected": -7.412909507751465, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.8530774321641297, | |
| "grad_norm": 6.305134676239398, | |
| "learning_rate": 7.785219433859846e-09, | |
| "logits/chosen": -2.666156768798828, | |
| "logits/rejected": -2.6787142753601074, | |
| "logps/chosen": -795.4716186523438, | |
| "logps/rejected": -1216.070068359375, | |
| "loss": 0.1889, | |
| "rewards/accuracies": 0.9312499761581421, | |
| "rewards/chosen": -3.1992480754852295, | |
| "rewards/margins": 4.258774757385254, | |
| "rewards/rejected": -7.4580230712890625, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.8795499669093316, | |
| "grad_norm": 8.35714287832233, | |
| "learning_rate": 5.177866543990689e-09, | |
| "logits/chosen": -2.6713449954986572, | |
| "logits/rejected": -2.7099664211273193, | |
| "logps/chosen": -809.7697143554688, | |
| "logps/rejected": -1260.0615234375, | |
| "loss": 0.1768, | |
| "rewards/accuracies": 0.925000011920929, | |
| "rewards/chosen": -3.31097412109375, | |
| "rewards/margins": 4.475188255310059, | |
| "rewards/rejected": -7.786163330078125, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.9060225016545336, | |
| "grad_norm": 6.357842566297534, | |
| "learning_rate": 3.0960627949644105e-09, | |
| "logits/chosen": -2.661090850830078, | |
| "logits/rejected": -2.6854565143585205, | |
| "logps/chosen": -799.8544921875, | |
| "logps/rejected": -1197.60595703125, | |
| "loss": 0.195, | |
| "rewards/accuracies": 0.9125000238418579, | |
| "rewards/chosen": -3.179225206375122, | |
| "rewards/margins": 4.040897846221924, | |
| "rewards/rejected": -7.220122337341309, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.9324950363997353, | |
| "grad_norm": 6.897936781273812, | |
| "learning_rate": 1.5442771053230663e-09, | |
| "logits/chosen": -2.654930353164673, | |
| "logits/rejected": -2.6900641918182373, | |
| "logps/chosen": -794.2493286132812, | |
| "logps/rejected": -1218.8597412109375, | |
| "loss": 0.1752, | |
| "rewards/accuracies": 0.925000011920929, | |
| "rewards/chosen": -3.182406187057495, | |
| "rewards/margins": 4.386531352996826, | |
| "rewards/rejected": -7.5689377784729, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.958967571144937, | |
| "grad_norm": 6.958740857504488, | |
| "learning_rate": 5.25840626643681e-10, | |
| "logits/chosen": -2.731703519821167, | |
| "logits/rejected": -2.728083848953247, | |
| "logps/chosen": -804.791259765625, | |
| "logps/rejected": -1165.920654296875, | |
| "loss": 0.1871, | |
| "rewards/accuracies": 0.8968750238418579, | |
| "rewards/chosen": -3.141540765762329, | |
| "rewards/margins": 3.8939075469970703, | |
| "rewards/rejected": -7.035449028015137, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.985440105890139, | |
| "grad_norm": 7.712958898241185, | |
| "learning_rate": 4.293959269863201e-11, | |
| "logits/chosen": -2.6934590339660645, | |
| "logits/rejected": -2.702648639678955, | |
| "logps/chosen": -818.2516479492188, | |
| "logps/rejected": -1221.8408203125, | |
| "loss": 0.2042, | |
| "rewards/accuracies": 0.925000011920929, | |
| "rewards/chosen": -3.207921266555786, | |
| "rewards/margins": 4.332727909088135, | |
| "rewards/rejected": -7.5406494140625, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.9960291197882198, | |
| "step": 754, | |
| "total_flos": 0.0, | |
| "train_loss": 0.26563876598520053, | |
| "train_runtime": 5267.2293, | |
| "train_samples_per_second": 36.699, | |
| "train_steps_per_second": 0.143 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 754, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |