| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 1250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 4e-08, | |
| "logits/chosen": 0.17149439454078674, | |
| "logits/rejected": -0.16577354073524475, | |
| "logps/chosen": -424.0601806640625, | |
| "logps/rejected": -451.33026123046875, | |
| "loss": 0.21, | |
| "rewards/accuracies": 0.0, | |
| "rewards/chosen": 0.0, | |
| "rewards/margins": 0.0, | |
| "rewards/rejected": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 4.0000000000000003e-07, | |
| "logits/chosen": 0.21744126081466675, | |
| "logits/rejected": 0.10579843074083328, | |
| "logps/chosen": -382.20452880859375, | |
| "logps/rejected": -405.85113525390625, | |
| "loss": 0.1999, | |
| "rewards/accuracies": 0.3194444477558136, | |
| "rewards/chosen": -0.000663464015815407, | |
| "rewards/margins": 2.6586050807964057e-05, | |
| "rewards/rejected": -0.0006900500156916678, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 8.000000000000001e-07, | |
| "logits/chosen": 0.09920214116573334, | |
| "logits/rejected": 0.17896269261837006, | |
| "logps/chosen": -424.150390625, | |
| "logps/rejected": -425.36322021484375, | |
| "loss": 0.2149, | |
| "rewards/accuracies": 0.3291666805744171, | |
| "rewards/chosen": -0.001120259752497077, | |
| "rewards/margins": -2.1472937078215182e-05, | |
| "rewards/rejected": -0.0010987869463860989, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.2000000000000002e-06, | |
| "logits/chosen": 0.10742081701755524, | |
| "logits/rejected": 0.17089782655239105, | |
| "logps/chosen": -428.20599365234375, | |
| "logps/rejected": -377.82623291015625, | |
| "loss": 0.2248, | |
| "rewards/accuracies": 0.32083335518836975, | |
| "rewards/chosen": -0.0012378226965665817, | |
| "rewards/margins": -4.129592707613483e-05, | |
| "rewards/rejected": -0.001196526805870235, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.6000000000000001e-06, | |
| "logits/chosen": 0.06296560913324356, | |
| "logits/rejected": 0.1074155792593956, | |
| "logps/chosen": -383.998291015625, | |
| "logps/rejected": -394.5549621582031, | |
| "loss": 0.2149, | |
| "rewards/accuracies": 0.3583333492279053, | |
| "rewards/chosen": -0.0029038754291832447, | |
| "rewards/margins": 4.263871960574761e-05, | |
| "rewards/rejected": -0.002946514170616865, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "logits/chosen": 0.14813974499702454, | |
| "logits/rejected": 0.09698332846164703, | |
| "logps/chosen": -429.5348205566406, | |
| "logps/rejected": -416.16082763671875, | |
| "loss": 0.2107, | |
| "rewards/accuracies": 0.3916666507720947, | |
| "rewards/chosen": -0.003114475170150399, | |
| "rewards/margins": 0.00012498181604314595, | |
| "rewards/rejected": -0.003239457029849291, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 2.4000000000000003e-06, | |
| "logits/chosen": 0.07653869688510895, | |
| "logits/rejected": 0.21269424259662628, | |
| "logps/chosen": -426.544189453125, | |
| "logps/rejected": -410.603271484375, | |
| "loss": 0.2157, | |
| "rewards/accuracies": 0.3916667103767395, | |
| "rewards/chosen": -0.003012719564139843, | |
| "rewards/margins": 4.8593028623145074e-05, | |
| "rewards/rejected": -0.003061312483623624, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.8000000000000003e-06, | |
| "logits/chosen": 0.01920640841126442, | |
| "logits/rejected": 0.16419027745723724, | |
| "logps/chosen": -390.6888122558594, | |
| "logps/rejected": -365.8466796875, | |
| "loss": 0.204, | |
| "rewards/accuracies": 0.38333332538604736, | |
| "rewards/chosen": -0.003003176301717758, | |
| "rewards/margins": 0.00020479671366047114, | |
| "rewards/rejected": -0.00320797273889184, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 3.2000000000000003e-06, | |
| "logits/chosen": 0.10941372811794281, | |
| "logits/rejected": 0.16332782804965973, | |
| "logps/chosen": -412.390625, | |
| "logps/rejected": -417.4317321777344, | |
| "loss": 0.2145, | |
| "rewards/accuracies": 0.4375, | |
| "rewards/chosen": -0.0038808684330433607, | |
| "rewards/margins": 0.0004434056463651359, | |
| "rewards/rejected": -0.004324274603277445, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "logits/chosen": 0.18084049224853516, | |
| "logits/rejected": 0.13003312051296234, | |
| "logps/chosen": -419.4568786621094, | |
| "logps/rejected": -383.35205078125, | |
| "loss": 0.2071, | |
| "rewards/accuracies": 0.42500001192092896, | |
| "rewards/chosen": -0.003477463498711586, | |
| "rewards/margins": 0.0005018340307287872, | |
| "rewards/rejected": -0.003979298286139965, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.000000000000001e-06, | |
| "logits/chosen": 0.13773226737976074, | |
| "logits/rejected": 0.136666402220726, | |
| "logps/chosen": -423.04364013671875, | |
| "logps/rejected": -416.84002685546875, | |
| "loss": 0.199, | |
| "rewards/accuracies": 0.4333333373069763, | |
| "rewards/chosen": -0.0012789195170626044, | |
| "rewards/margins": 0.00046353376819752157, | |
| "rewards/rejected": -0.0017424530815333128, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.4e-06, | |
| "logits/chosen": 0.07242347300052643, | |
| "logits/rejected": 0.10395532846450806, | |
| "logps/chosen": -410.901611328125, | |
| "logps/rejected": -425.29461669921875, | |
| "loss": 0.2097, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.0012211017310619354, | |
| "rewards/margins": 0.0012917147250846028, | |
| "rewards/rejected": -0.0025128168053925037, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.800000000000001e-06, | |
| "logits/chosen": 0.16134211421012878, | |
| "logits/rejected": -0.009699034504592419, | |
| "logps/chosen": -399.96441650390625, | |
| "logps/rejected": -396.57672119140625, | |
| "loss": 0.2075, | |
| "rewards/accuracies": 0.4958333373069763, | |
| "rewards/chosen": -0.0006208900013007224, | |
| "rewards/margins": 0.0016982462257146835, | |
| "rewards/rejected": -0.002319136168807745, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.999756310023261e-06, | |
| "logits/chosen": 0.07792703807353973, | |
| "logits/rejected": 0.0053740618750452995, | |
| "logps/chosen": -407.54522705078125, | |
| "logps/rejected": -386.1396179199219, | |
| "loss": 0.2103, | |
| "rewards/accuracies": 0.49166664481163025, | |
| "rewards/chosen": 0.0015033172676339746, | |
| "rewards/margins": 0.0024497557897120714, | |
| "rewards/rejected": -0.0009464388713240623, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.997807075247147e-06, | |
| "logits/chosen": 0.17738033831119537, | |
| "logits/rejected": 0.11884935200214386, | |
| "logps/chosen": -432.95916748046875, | |
| "logps/rejected": -437.46685791015625, | |
| "loss": 0.2088, | |
| "rewards/accuracies": 0.5041666626930237, | |
| "rewards/chosen": 0.008168506436049938, | |
| "rewards/margins": 0.0032126710284501314, | |
| "rewards/rejected": 0.0049558356404304504, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 4.993910125649561e-06, | |
| "logits/chosen": 0.09095987677574158, | |
| "logits/rejected": 0.21810762584209442, | |
| "logps/chosen": -403.97509765625, | |
| "logps/rejected": -398.00567626953125, | |
| "loss": 0.214, | |
| "rewards/accuracies": 0.4833332896232605, | |
| "rewards/chosen": 0.012083831243216991, | |
| "rewards/margins": 0.002966669388115406, | |
| "rewards/rejected": 0.009117161855101585, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.988068499954578e-06, | |
| "logits/chosen": 0.11167311668395996, | |
| "logits/rejected": -0.01715979352593422, | |
| "logps/chosen": -395.1109924316406, | |
| "logps/rejected": -386.7571105957031, | |
| "loss": 0.2005, | |
| "rewards/accuracies": 0.4791666567325592, | |
| "rewards/chosen": 0.01772935874760151, | |
| "rewards/margins": 0.005436168517917395, | |
| "rewards/rejected": 0.012293190695345402, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.980286753286196e-06, | |
| "logits/chosen": 0.07135671377182007, | |
| "logits/rejected": 0.004518161062151194, | |
| "logps/chosen": -381.22509765625, | |
| "logps/rejected": -369.58880615234375, | |
| "loss": 0.2085, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": 0.027819514274597168, | |
| "rewards/margins": 0.006179440300911665, | |
| "rewards/rejected": 0.021640075370669365, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 4.970570953616383e-06, | |
| "logits/chosen": 0.03489253669977188, | |
| "logits/rejected": 0.12006882578134537, | |
| "logps/chosen": -377.1296081542969, | |
| "logps/rejected": -362.04364013671875, | |
| "loss": 0.2018, | |
| "rewards/accuracies": 0.5166667103767395, | |
| "rewards/chosen": 0.029645437374711037, | |
| "rewards/margins": 0.008070086129009724, | |
| "rewards/rejected": 0.02157535031437874, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 4.958928677033465e-06, | |
| "logits/chosen": 0.01938822492957115, | |
| "logits/rejected": 0.04349769651889801, | |
| "logps/chosen": -394.98101806640625, | |
| "logps/rejected": -428.86248779296875, | |
| "loss": 0.208, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": 0.020193781703710556, | |
| "rewards/margins": 0.008541513234376907, | |
| "rewards/rejected": 0.011652267538011074, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.9453690018345144e-06, | |
| "logits/chosen": -0.11528314650058746, | |
| "logits/rejected": 0.08821746706962585, | |
| "logps/chosen": -394.2080078125, | |
| "logps/rejected": -403.65496826171875, | |
| "loss": 0.1946, | |
| "rewards/accuracies": 0.5250000357627869, | |
| "rewards/chosen": 0.020861472934484482, | |
| "rewards/margins": 0.01827095076441765, | |
| "rewards/rejected": 0.0025905228685587645, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 4.9299025014463665e-06, | |
| "logits/chosen": -0.08552906662225723, | |
| "logits/rejected": 0.031030725687742233, | |
| "logps/chosen": -383.574462890625, | |
| "logps/rejected": -404.0837707519531, | |
| "loss": 0.1909, | |
| "rewards/accuracies": 0.47083336114883423, | |
| "rewards/chosen": 0.01809662953019142, | |
| "rewards/margins": 0.014171933755278587, | |
| "rewards/rejected": 0.0039246962405741215, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.912541236180779e-06, | |
| "logits/chosen": -0.13254012167453766, | |
| "logits/rejected": -0.06545063108205795, | |
| "logps/chosen": -332.7752380371094, | |
| "logps/rejected": -348.6072082519531, | |
| "loss": 0.1975, | |
| "rewards/accuracies": 0.44583338499069214, | |
| "rewards/chosen": 0.0345706008374691, | |
| "rewards/margins": 0.018852438777685165, | |
| "rewards/rejected": 0.015718165785074234, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.893298743830168e-06, | |
| "logits/chosen": -0.09530626982450485, | |
| "logits/rejected": -0.10254266113042831, | |
| "logps/chosen": -398.1355895996094, | |
| "logps/rejected": -425.45172119140625, | |
| "loss": 0.1885, | |
| "rewards/accuracies": 0.47083336114883423, | |
| "rewards/chosen": 0.012728950008749962, | |
| "rewards/margins": 0.029127534478902817, | |
| "rewards/rejected": -0.016398586332798004, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.8721900291112415e-06, | |
| "logits/chosen": -0.13877639174461365, | |
| "logits/rejected": 0.013827897608280182, | |
| "logps/chosen": -396.7447204589844, | |
| "logps/rejected": -418.518798828125, | |
| "loss": 0.1898, | |
| "rewards/accuracies": 0.5333333611488342, | |
| "rewards/chosen": -0.002779316157102585, | |
| "rewards/margins": 0.028926661238074303, | |
| "rewards/rejected": -0.03170597925782204, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.849231551964771e-06, | |
| "logits/chosen": -0.09128137677907944, | |
| "logits/rejected": -0.05126717686653137, | |
| "logps/chosen": -404.491943359375, | |
| "logps/rejected": -454.119873046875, | |
| "loss": 0.1884, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.019060546532273293, | |
| "rewards/margins": 0.036740098148584366, | |
| "rewards/rejected": -0.05580064654350281, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 4.824441214720629e-06, | |
| "logits/chosen": -0.11080489307641983, | |
| "logits/rejected": -0.14405983686447144, | |
| "logps/chosen": -494.0188903808594, | |
| "logps/rejected": -519.1729736328125, | |
| "loss": 0.2014, | |
| "rewards/accuracies": 0.5250000357627869, | |
| "rewards/chosen": -0.0692259818315506, | |
| "rewards/margins": 0.04139824956655502, | |
| "rewards/rejected": -0.11062423884868622, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.7978383481380865e-06, | |
| "logits/chosen": -0.12256069481372833, | |
| "logits/rejected": -0.13387584686279297, | |
| "logps/chosen": -487.1686096191406, | |
| "logps/rejected": -525.7955322265625, | |
| "loss": 0.2103, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.09114581346511841, | |
| "rewards/margins": 0.04267682135105133, | |
| "rewards/rejected": -0.13382264971733093, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.769443696332272e-06, | |
| "logits/chosen": -0.049383699893951416, | |
| "logits/rejected": -0.16224342584609985, | |
| "logps/chosen": -455.2054748535156, | |
| "logps/rejected": -494.6611328125, | |
| "loss": 0.1946, | |
| "rewards/accuracies": 0.429166704416275, | |
| "rewards/chosen": -0.08516400307416916, | |
| "rewards/margins": 0.03522291034460068, | |
| "rewards/rejected": -0.12038691341876984, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.7392794005985324e-06, | |
| "logits/chosen": -0.08826161921024323, | |
| "logits/rejected": -0.05565328150987625, | |
| "logps/chosen": -510.9239196777344, | |
| "logps/rejected": -519.4769897460938, | |
| "loss": 0.202, | |
| "rewards/accuracies": 0.4750000536441803, | |
| "rewards/chosen": -0.10056551545858383, | |
| "rewards/margins": 0.028566282242536545, | |
| "rewards/rejected": -0.12913179397583008, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.707368982147318e-06, | |
| "logits/chosen": 0.010274168103933334, | |
| "logits/rejected": -0.10608565807342529, | |
| "logps/chosen": -440.32464599609375, | |
| "logps/rejected": -487.58209228515625, | |
| "loss": 0.1839, | |
| "rewards/accuracies": 0.4750000536441803, | |
| "rewards/chosen": -0.052764929831027985, | |
| "rewards/margins": 0.04310880973935127, | |
| "rewards/rejected": -0.09587373584508896, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 4.673737323763048e-06, | |
| "logits/chosen": -0.016120150685310364, | |
| "logits/rejected": 0.003306228667497635, | |
| "logps/chosen": -387.26629638671875, | |
| "logps/rejected": -432.75018310546875, | |
| "loss": 0.1978, | |
| "rewards/accuracies": 0.44166669249534607, | |
| "rewards/chosen": -0.039925262331962585, | |
| "rewards/margins": 0.03133027255535126, | |
| "rewards/rejected": -0.07125553488731384, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.638410650401267e-06, | |
| "logits/chosen": -0.036097295582294464, | |
| "logits/rejected": -0.040274590253829956, | |
| "logps/chosen": -436.0054626464844, | |
| "logps/rejected": -477.3866271972656, | |
| "loss": 0.1853, | |
| "rewards/accuracies": 0.49166664481163025, | |
| "rewards/chosen": -0.04764013737440109, | |
| "rewards/margins": 0.02985115721821785, | |
| "rewards/rejected": -0.07749129831790924, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.601416508739211e-06, | |
| "logits/chosen": -0.0325358584523201, | |
| "logits/rejected": -0.036110930144786835, | |
| "logps/chosen": -475.7208557128906, | |
| "logps/rejected": -515.5665283203125, | |
| "loss": 0.1944, | |
| "rewards/accuracies": 0.5208333730697632, | |
| "rewards/chosen": -0.04646621271967888, | |
| "rewards/margins": 0.04523799195885658, | |
| "rewards/rejected": -0.09170420467853546, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.562783745695738e-06, | |
| "logits/chosen": -0.04594831541180611, | |
| "logits/rejected": -0.03698522225022316, | |
| "logps/chosen": -496.4822692871094, | |
| "logps/rejected": -489.32391357421875, | |
| "loss": 0.2003, | |
| "rewards/accuracies": 0.47083336114883423, | |
| "rewards/chosen": -0.05407736450433731, | |
| "rewards/margins": 0.03637322038412094, | |
| "rewards/rejected": -0.09045058488845825, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 4.522542485937369e-06, | |
| "logits/chosen": -0.1583801805973053, | |
| "logits/rejected": -0.03529781848192215, | |
| "logps/chosen": -460.81793212890625, | |
| "logps/rejected": -519.77587890625, | |
| "loss": 0.1752, | |
| "rewards/accuracies": 0.5291666984558105, | |
| "rewards/chosen": -0.03841938078403473, | |
| "rewards/margins": 0.0512566938996315, | |
| "rewards/rejected": -0.08967608213424683, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.4807241083879774e-06, | |
| "logits/chosen": -0.20257849991321564, | |
| "logits/rejected": -0.1146155446767807, | |
| "logps/chosen": -442.28594970703125, | |
| "logps/rejected": -458.53173828125, | |
| "loss": 0.1895, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.04586230590939522, | |
| "rewards/margins": 0.03675750643014908, | |
| "rewards/rejected": -0.0826198160648346, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.437361221760449e-06, | |
| "logits/chosen": -0.11894889175891876, | |
| "logits/rejected": -0.08775348216295242, | |
| "logps/chosen": -456.0023498535156, | |
| "logps/rejected": -466.5443420410156, | |
| "loss": 0.1868, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.050740133970975876, | |
| "rewards/margins": 0.039192140102386475, | |
| "rewards/rejected": -0.08993227034807205, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 4.3924876391293915e-06, | |
| "logits/chosen": -0.18518304824829102, | |
| "logits/rejected": -0.1563677340745926, | |
| "logps/chosen": -445.9903259277344, | |
| "logps/rejected": -468.9036560058594, | |
| "loss": 0.1779, | |
| "rewards/accuracies": 0.5000000596046448, | |
| "rewards/chosen": -0.047802653163671494, | |
| "rewards/margins": 0.03556705266237259, | |
| "rewards/rejected": -0.08336970955133438, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.346138351564711e-06, | |
| "logits/chosen": -0.17687319219112396, | |
| "logits/rejected": -0.16852393746376038, | |
| "logps/chosen": -413.527587890625, | |
| "logps/rejected": -469.2893981933594, | |
| "loss": 0.1849, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.047930918633937836, | |
| "rewards/margins": 0.05219081789255142, | |
| "rewards/rejected": -0.10012175142765045, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.2983495008466285e-06, | |
| "logits/chosen": -0.20533815026283264, | |
| "logits/rejected": -0.20098014175891876, | |
| "logps/chosen": -412.97698974609375, | |
| "logps/rejected": -473.263671875, | |
| "loss": 0.1845, | |
| "rewards/accuracies": 0.4958333373069763, | |
| "rewards/chosen": -0.04717981815338135, | |
| "rewards/margins": 0.05400770157575607, | |
| "rewards/rejected": -0.10118752717971802, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.249158351283414e-06, | |
| "logits/chosen": -0.20028965175151825, | |
| "logits/rejected": -0.11055328696966171, | |
| "logps/chosen": -464.6874084472656, | |
| "logps/rejected": -541.6143798828125, | |
| "loss": 0.1788, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.05910419300198555, | |
| "rewards/margins": 0.060402512550354004, | |
| "rewards/rejected": -0.11950669437646866, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.198603260653792e-06, | |
| "logits/chosen": -0.23082371056079865, | |
| "logits/rejected": -0.12578482925891876, | |
| "logps/chosen": -482.4796447753906, | |
| "logps/rejected": -553.0191650390625, | |
| "loss": 0.1709, | |
| "rewards/accuracies": 0.5166667103767395, | |
| "rewards/chosen": -0.058818310499191284, | |
| "rewards/margins": 0.06956759095191956, | |
| "rewards/rejected": -0.12838590145111084, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.146723650296701e-06, | |
| "logits/chosen": -0.1750180423259735, | |
| "logits/rejected": -0.11045924574136734, | |
| "logps/chosen": -482.70904541015625, | |
| "logps/rejected": -499.7896423339844, | |
| "loss": 0.187, | |
| "rewards/accuracies": 0.46666663885116577, | |
| "rewards/chosen": -0.0458199642598629, | |
| "rewards/margins": 0.04604263976216316, | |
| "rewards/rejected": -0.09186260402202606, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.093559974371725e-06, | |
| "logits/chosen": -0.2370443344116211, | |
| "logits/rejected": -0.14606305956840515, | |
| "logps/chosen": -397.68548583984375, | |
| "logps/rejected": -440.5757751464844, | |
| "loss": 0.1907, | |
| "rewards/accuracies": 0.47083336114883423, | |
| "rewards/chosen": -0.02382233552634716, | |
| "rewards/margins": 0.04170426353812218, | |
| "rewards/rejected": -0.06552660465240479, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.039153688314146e-06, | |
| "logits/chosen": -0.19962234795093536, | |
| "logits/rejected": -0.18455585837364197, | |
| "logps/chosen": -433.9319763183594, | |
| "logps/rejected": -486.72247314453125, | |
| "loss": 0.1899, | |
| "rewards/accuracies": 0.5333333611488342, | |
| "rewards/chosen": -0.005404153373092413, | |
| "rewards/margins": 0.050141505897045135, | |
| "rewards/rejected": -0.05554566532373428, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 3.983547216509254e-06, | |
| "logits/chosen": -0.23652057349681854, | |
| "logits/rejected": -0.13747069239616394, | |
| "logps/chosen": -385.99859619140625, | |
| "logps/rejected": -418.61859130859375, | |
| "loss": 0.1786, | |
| "rewards/accuracies": 0.47083336114883423, | |
| "rewards/chosen": -0.011283977888524532, | |
| "rewards/margins": 0.043274398893117905, | |
| "rewards/rejected": -0.05455837398767471, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.92678391921108e-06, | |
| "logits/chosen": -0.1722879260778427, | |
| "logits/rejected": -0.2078588902950287, | |
| "logps/chosen": -425.4927673339844, | |
| "logps/rejected": -452.87347412109375, | |
| "loss": 0.1813, | |
| "rewards/accuracies": 0.5041666626930237, | |
| "rewards/chosen": -0.0162376556545496, | |
| "rewards/margins": 0.04304374381899834, | |
| "rewards/rejected": -0.059281397610902786, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 3.868908058731376e-06, | |
| "logits/chosen": -0.21139149367809296, | |
| "logits/rejected": -0.17750591039657593, | |
| "logps/chosen": -478.3169860839844, | |
| "logps/rejected": -513.8565063476562, | |
| "loss": 0.1842, | |
| "rewards/accuracies": 0.5333333611488342, | |
| "rewards/chosen": -0.023320624604821205, | |
| "rewards/margins": 0.044868770986795425, | |
| "rewards/rejected": -0.06818939745426178, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 3.8099647649251984e-06, | |
| "logits/chosen": -0.2609555125236511, | |
| "logits/rejected": -0.17367315292358398, | |
| "logps/chosen": -418.36279296875, | |
| "logps/rejected": -473.0401306152344, | |
| "loss": 0.1892, | |
| "rewards/accuracies": 0.4958333373069763, | |
| "rewards/chosen": -0.032408542931079865, | |
| "rewards/margins": 0.04911263287067413, | |
| "rewards/rejected": -0.0815211683511734, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "logits/chosen": -0.2730502486228943, | |
| "logits/rejected": -0.2116207778453827, | |
| "logps/chosen": -475.0166015625, | |
| "logps/rejected": -491.03948974609375, | |
| "loss": 0.1987, | |
| "rewards/accuracies": 0.4791666567325592, | |
| "rewards/chosen": -0.04057503864169121, | |
| "rewards/margins": 0.039973508566617966, | |
| "rewards/rejected": -0.08054854720830917, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3.689060522675689e-06, | |
| "logits/chosen": -0.2958037555217743, | |
| "logits/rejected": -0.16685570776462555, | |
| "logps/chosen": -446.9232482910156, | |
| "logps/rejected": -492.4410095214844, | |
| "loss": 0.182, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.04383234679698944, | |
| "rewards/margins": 0.04704504832625389, | |
| "rewards/rejected": -0.09087739884853363, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3.627193851723577e-06, | |
| "logits/chosen": -0.21801650524139404, | |
| "logits/rejected": -0.23087890446186066, | |
| "logps/chosen": -416.719482421875, | |
| "logps/rejected": -477.15155029296875, | |
| "loss": 0.1968, | |
| "rewards/accuracies": 0.4791666567325592, | |
| "rewards/chosen": -0.041071660816669464, | |
| "rewards/margins": 0.04972352460026741, | |
| "rewards/rejected": -0.09079518169164658, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 3.564448228912682e-06, | |
| "logits/chosen": -0.22313952445983887, | |
| "logits/rejected": -0.2395196259021759, | |
| "logps/chosen": -435.31829833984375, | |
| "logps/rejected": -500.8714294433594, | |
| "loss": 0.1991, | |
| "rewards/accuracies": 0.5083333253860474, | |
| "rewards/chosen": -0.050644002854824066, | |
| "rewards/margins": 0.04440879076719284, | |
| "rewards/rejected": -0.0950528010725975, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 3.5008725813922383e-06, | |
| "logits/chosen": -0.29003390669822693, | |
| "logits/rejected": -0.15861546993255615, | |
| "logps/chosen": -495.62060546875, | |
| "logps/rejected": -535.9207763671875, | |
| "loss": 0.1652, | |
| "rewards/accuracies": 0.5666667222976685, | |
| "rewards/chosen": -0.047018229961395264, | |
| "rewards/margins": 0.04982719197869301, | |
| "rewards/rejected": -0.09684541821479797, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3.436516483539781e-06, | |
| "logits/chosen": -0.28255754709243774, | |
| "logits/rejected": -0.2905382513999939, | |
| "logps/chosen": -474.9947204589844, | |
| "logps/rejected": -492.84130859375, | |
| "loss": 0.1953, | |
| "rewards/accuracies": 0.48750004172325134, | |
| "rewards/chosen": -0.06080613657832146, | |
| "rewards/margins": 0.035242706537246704, | |
| "rewards/rejected": -0.09604884684085846, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 3.3714301183045382e-06, | |
| "logits/chosen": -0.2140432894229889, | |
| "logits/rejected": -0.30546998977661133, | |
| "logps/chosen": -459.7528381347656, | |
| "logps/rejected": -547.1843872070312, | |
| "loss": 0.1525, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.058339886367321014, | |
| "rewards/margins": 0.07008499652147293, | |
| "rewards/rejected": -0.12842489778995514, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 3.3056642380762783e-06, | |
| "logits/chosen": -0.20786988735198975, | |
| "logits/rejected": -0.30285245180130005, | |
| "logps/chosen": -435.9219665527344, | |
| "logps/rejected": -506.14996337890625, | |
| "loss": 0.1704, | |
| "rewards/accuracies": 0.49583330750465393, | |
| "rewards/chosen": -0.06140541285276413, | |
| "rewards/margins": 0.0633297935128212, | |
| "rewards/rejected": -0.12473521381616592, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 3.2392701251101172e-06, | |
| "logits/chosen": -0.2695538401603699, | |
| "logits/rejected": -0.2578602135181427, | |
| "logps/chosen": -507.4090881347656, | |
| "logps/rejected": -573.2657470703125, | |
| "loss": 0.1713, | |
| "rewards/accuracies": 0.5583332777023315, | |
| "rewards/chosen": -0.07743930071592331, | |
| "rewards/margins": 0.0685843974351883, | |
| "rewards/rejected": -0.1460237056016922, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 3.1722995515381644e-06, | |
| "logits/chosen": -0.2915573716163635, | |
| "logits/rejected": -0.2425670623779297, | |
| "logps/chosen": -462.32220458984375, | |
| "logps/rejected": -534.523193359375, | |
| "loss": 0.1851, | |
| "rewards/accuracies": 0.5041666626930237, | |
| "rewards/chosen": -0.07182817161083221, | |
| "rewards/margins": 0.06546822935342789, | |
| "rewards/rejected": -0.1372964084148407, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 3.1048047389991693e-06, | |
| "logits/chosen": -0.2364952117204666, | |
| "logits/rejected": -0.29197296500205994, | |
| "logps/chosen": -456.0204162597656, | |
| "logps/rejected": -503.23822021484375, | |
| "loss": 0.1863, | |
| "rewards/accuracies": 0.48750004172325134, | |
| "rewards/chosen": -0.06296292692422867, | |
| "rewards/margins": 0.05075465515255928, | |
| "rewards/rejected": -0.11371759325265884, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.0368383179176584e-06, | |
| "logits/chosen": -0.3178825378417969, | |
| "logits/rejected": -0.17681264877319336, | |
| "logps/chosen": -445.4483947753906, | |
| "logps/rejected": -500.03387451171875, | |
| "loss": 0.1853, | |
| "rewards/accuracies": 0.4791666567325592, | |
| "rewards/chosen": -0.05490962415933609, | |
| "rewards/margins": 0.047073423862457275, | |
| "rewards/rejected": -0.10198304802179337, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.9684532864643123e-06, | |
| "logits/chosen": -0.3539500832557678, | |
| "logits/rejected": -0.20667648315429688, | |
| "logps/chosen": -466.39764404296875, | |
| "logps/rejected": -501.822021484375, | |
| "loss": 0.1844, | |
| "rewards/accuracies": 0.4833333492279053, | |
| "rewards/chosen": -0.05339791625738144, | |
| "rewards/margins": 0.053281139582395554, | |
| "rewards/rejected": -0.1066790446639061, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2.8997029692295875e-06, | |
| "logits/chosen": -0.3101775050163269, | |
| "logits/rejected": -0.2914181053638458, | |
| "logps/chosen": -427.52032470703125, | |
| "logps/rejected": -466.09197998046875, | |
| "loss": 0.1837, | |
| "rewards/accuracies": 0.4749999940395355, | |
| "rewards/chosen": -0.048539165407419205, | |
| "rewards/margins": 0.052506644278764725, | |
| "rewards/rejected": -0.10104580223560333, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 2.8306409756428067e-06, | |
| "logits/chosen": -0.317528635263443, | |
| "logits/rejected": -0.18300285935401917, | |
| "logps/chosen": -482.93939208984375, | |
| "logps/rejected": -531.2368774414062, | |
| "loss": 0.1641, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.041467417031526566, | |
| "rewards/margins": 0.07688767462968826, | |
| "rewards/rejected": -0.11835509538650513, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 2.761321158169134e-06, | |
| "logits/chosen": -0.23212091624736786, | |
| "logits/rejected": -0.28510648012161255, | |
| "logps/chosen": -484.9613342285156, | |
| "logps/rejected": -542.4539184570312, | |
| "loss": 0.1745, | |
| "rewards/accuracies": 0.5250000357627869, | |
| "rewards/chosen": -0.039610255509614944, | |
| "rewards/margins": 0.061538826674222946, | |
| "rewards/rejected": -0.10114908218383789, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 2.6917975703170466e-06, | |
| "logits/chosen": -0.27057233452796936, | |
| "logits/rejected": -0.24964137375354767, | |
| "logps/chosen": -427.0650329589844, | |
| "logps/rejected": -500.9590759277344, | |
| "loss": 0.1835, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.0389983169734478, | |
| "rewards/margins": 0.07033900916576385, | |
| "rewards/rejected": -0.10933732986450195, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.6221244244890336e-06, | |
| "logits/chosen": -0.2637156546115875, | |
| "logits/rejected": -0.2788185477256775, | |
| "logps/chosen": -450.2159118652344, | |
| "logps/rejected": -507.42962646484375, | |
| "loss": 0.1696, | |
| "rewards/accuracies": 0.4958333969116211, | |
| "rewards/chosen": -0.0435161329805851, | |
| "rewards/margins": 0.0677158460021019, | |
| "rewards/rejected": -0.1112319827079773, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.5523560497083927e-06, | |
| "logits/chosen": -0.23590517044067383, | |
| "logits/rejected": -0.30847400426864624, | |
| "logps/chosen": -388.68463134765625, | |
| "logps/rejected": -449.8109436035156, | |
| "loss": 0.1846, | |
| "rewards/accuracies": 0.4958333373069763, | |
| "rewards/chosen": -0.033804431557655334, | |
| "rewards/margins": 0.06065298244357109, | |
| "rewards/rejected": -0.09445741027593613, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.482546849255096e-06, | |
| "logits/chosen": -0.310326486825943, | |
| "logits/rejected": -0.2718106508255005, | |
| "logps/chosen": -451.0816345214844, | |
| "logps/rejected": -527.0636596679688, | |
| "loss": 0.18, | |
| "rewards/accuracies": 0.5125000476837158, | |
| "rewards/chosen": -0.04315061867237091, | |
| "rewards/margins": 0.06889710575342178, | |
| "rewards/rejected": -0.1120477169752121, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 2.4127512582437486e-06, | |
| "logits/chosen": -0.2296207696199417, | |
| "logits/rejected": -0.2563227415084839, | |
| "logps/chosen": -489.15380859375, | |
| "logps/rejected": -535.2841186523438, | |
| "loss": 0.1847, | |
| "rewards/accuracies": 0.5333333611488342, | |
| "rewards/chosen": -0.04005949944257736, | |
| "rewards/margins": 0.063002809882164, | |
| "rewards/rejected": -0.10306230932474136, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 2.3430237011767166e-06, | |
| "logits/chosen": -0.28292739391326904, | |
| "logits/rejected": -0.22382238507270813, | |
| "logps/chosen": -402.1965637207031, | |
| "logps/rejected": -477.5860900878906, | |
| "loss": 0.1755, | |
| "rewards/accuracies": 0.5041667222976685, | |
| "rewards/chosen": -0.018582064658403397, | |
| "rewards/margins": 0.07370585948228836, | |
| "rewards/rejected": -0.09228792786598206, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.2734185495055503e-06, | |
| "logits/chosen": -0.286498486995697, | |
| "logits/rejected": -0.22381773591041565, | |
| "logps/chosen": -419.038330078125, | |
| "logps/rejected": -441.9375915527344, | |
| "loss": 0.1932, | |
| "rewards/accuracies": 0.44583335518836975, | |
| "rewards/chosen": -0.022441592067480087, | |
| "rewards/margins": 0.03786554932594299, | |
| "rewards/rejected": -0.06030714511871338, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 2.2039900792337477e-06, | |
| "logits/chosen": -0.22549033164978027, | |
| "logits/rejected": -0.2243853360414505, | |
| "logps/chosen": -447.57635498046875, | |
| "logps/rejected": -464.91082763671875, | |
| "loss": 0.1868, | |
| "rewards/accuracies": 0.5291666388511658, | |
| "rewards/chosen": -0.014703089371323586, | |
| "rewards/margins": 0.04347188025712967, | |
| "rewards/rejected": -0.058174967765808105, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 2.134792428593971e-06, | |
| "logits/chosen": -0.26262083649635315, | |
| "logits/rejected": -0.23271727561950684, | |
| "logps/chosen": -452.77880859375, | |
| "logps/rejected": -475.3125915527344, | |
| "loss": 0.1925, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.025184612721204758, | |
| "rewards/margins": 0.04363705962896347, | |
| "rewards/rejected": -0.06882167607545853, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.0658795558326745e-06, | |
| "logits/chosen": -0.2798984944820404, | |
| "logits/rejected": -0.1053474098443985, | |
| "logps/chosen": -432.2371520996094, | |
| "logps/rejected": -437.374755859375, | |
| "loss": 0.183, | |
| "rewards/accuracies": 0.4833333492279053, | |
| "rewards/chosen": -0.0092130983248353, | |
| "rewards/margins": 0.04416029527783394, | |
| "rewards/rejected": -0.053373388946056366, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.997305197135089e-06, | |
| "logits/chosen": -0.24332702159881592, | |
| "logits/rejected": -0.24699708819389343, | |
| "logps/chosen": -413.81317138671875, | |
| "logps/rejected": -436.09185791015625, | |
| "loss": 0.1861, | |
| "rewards/accuracies": 0.4833333492279053, | |
| "rewards/chosen": -0.018954753875732422, | |
| "rewards/margins": 0.046722229570150375, | |
| "rewards/rejected": -0.0656769797205925, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.9291228247233607e-06, | |
| "logits/chosen": -0.19670598208904266, | |
| "logits/rejected": -0.14346536993980408, | |
| "logps/chosen": -426.93316650390625, | |
| "logps/rejected": -422.44830322265625, | |
| "loss": 0.1813, | |
| "rewards/accuracies": 0.4833332896232605, | |
| "rewards/chosen": -0.01260639913380146, | |
| "rewards/margins": 0.04483729228377342, | |
| "rewards/rejected": -0.05744368955492973, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.8613856051605242e-06, | |
| "logits/chosen": -0.19106583297252655, | |
| "logits/rejected": -0.2591181695461273, | |
| "logps/chosen": -396.3172912597656, | |
| "logps/rejected": -458.3994140625, | |
| "loss": 0.1657, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.011414697393774986, | |
| "rewards/margins": 0.06354735046625137, | |
| "rewards/rejected": -0.07496203482151031, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.7941463578928088e-06, | |
| "logits/chosen": -0.20239433646202087, | |
| "logits/rejected": -0.18643388152122498, | |
| "logps/chosen": -386.0264587402344, | |
| "logps/rejected": -415.656494140625, | |
| "loss": 0.1868, | |
| "rewards/accuracies": 0.44583335518836975, | |
| "rewards/chosen": -0.01508850883692503, | |
| "rewards/margins": 0.044246602803468704, | |
| "rewards/rejected": -0.05933511257171631, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.7274575140626318e-06, | |
| "logits/chosen": -0.2549183666706085, | |
| "logits/rejected": -0.13016203045845032, | |
| "logps/chosen": -463.55816650390625, | |
| "logps/rejected": -502.923828125, | |
| "loss": 0.1769, | |
| "rewards/accuracies": 0.5583332777023315, | |
| "rewards/chosen": -0.025710273534059525, | |
| "rewards/margins": 0.05553792044520378, | |
| "rewards/rejected": -0.0812481939792633, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.661371075624363e-06, | |
| "logits/chosen": -0.2786555886268616, | |
| "logits/rejected": -0.19162516295909882, | |
| "logps/chosen": -460.2269592285156, | |
| "logps/rejected": -480.3688049316406, | |
| "loss": 0.1761, | |
| "rewards/accuracies": 0.5166667103767395, | |
| "rewards/chosen": -0.025559108704328537, | |
| "rewards/margins": 0.051572393625974655, | |
| "rewards/rejected": -0.0771314948797226, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.5959385747947697e-06, | |
| "logits/chosen": -0.16187801957130432, | |
| "logits/rejected": -0.15380509197711945, | |
| "logps/chosen": -437.10205078125, | |
| "logps/rejected": -482.42718505859375, | |
| "loss": 0.1891, | |
| "rewards/accuracies": 0.4958333969116211, | |
| "rewards/chosen": -0.02434079721570015, | |
| "rewards/margins": 0.04448997974395752, | |
| "rewards/rejected": -0.06883077323436737, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 1.5312110338697427e-06, | |
| "logits/chosen": -0.12476430833339691, | |
| "logits/rejected": -0.2756340801715851, | |
| "logps/chosen": -413.1475524902344, | |
| "logps/rejected": -469.5107421875, | |
| "loss": 0.1796, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.024911707267165184, | |
| "rewards/margins": 0.06037444993853569, | |
| "rewards/rejected": -0.08528616279363632, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.467238925438646e-06, | |
| "logits/chosen": -0.2615768313407898, | |
| "logits/rejected": -0.20825617015361786, | |
| "logps/chosen": -413.496826171875, | |
| "logps/rejected": -466.8892517089844, | |
| "loss": 0.1772, | |
| "rewards/accuracies": 0.5166667103767395, | |
| "rewards/chosen": -0.015650052577257156, | |
| "rewards/margins": 0.05704290792346001, | |
| "rewards/rejected": -0.07269296050071716, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.4040721330273063e-06, | |
| "logits/chosen": -0.23080775141716003, | |
| "logits/rejected": -0.07844565808773041, | |
| "logps/chosen": -432.4065856933594, | |
| "logps/rejected": -493.73876953125, | |
| "loss": 0.1733, | |
| "rewards/accuracies": 0.5208333134651184, | |
| "rewards/chosen": -0.02299593947827816, | |
| "rewards/margins": 0.06412651389837265, | |
| "rewards/rejected": -0.08712244778871536, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.3417599122003464e-06, | |
| "logits/chosen": -0.18598750233650208, | |
| "logits/rejected": -0.20719997584819794, | |
| "logps/chosen": -423.7872619628906, | |
| "logps/rejected": -479.3990173339844, | |
| "loss": 0.1958, | |
| "rewards/accuracies": 0.4791666567325592, | |
| "rewards/chosen": -0.027820762246847153, | |
| "rewards/margins": 0.05425436049699783, | |
| "rewards/rejected": -0.08207512646913528, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.280350852153168e-06, | |
| "logits/chosen": -0.2046211063861847, | |
| "logits/rejected": -0.23867718875408173, | |
| "logps/chosen": -432.0369567871094, | |
| "logps/rejected": -490.05548095703125, | |
| "loss": 0.1985, | |
| "rewards/accuracies": 0.4833333492279053, | |
| "rewards/chosen": -0.0314149484038353, | |
| "rewards/margins": 0.05088344216346741, | |
| "rewards/rejected": -0.0822983831167221, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.2198928378235717e-06, | |
| "logits/chosen": -0.2688402533531189, | |
| "logits/rejected": -0.20974984765052795, | |
| "logps/chosen": -416.24322509765625, | |
| "logps/rejected": -463.6009216308594, | |
| "loss": 0.1971, | |
| "rewards/accuracies": 0.46250003576278687, | |
| "rewards/chosen": -0.029970809817314148, | |
| "rewards/margins": 0.04484293609857559, | |
| "rewards/rejected": -0.07481373846530914, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.160433012552508e-06, | |
| "logits/chosen": -0.2599441409111023, | |
| "logits/rejected": -0.23637132346630096, | |
| "logps/chosen": -417.55908203125, | |
| "logps/rejected": -505.0204162597656, | |
| "loss": 0.1608, | |
| "rewards/accuracies": 0.5333333611488342, | |
| "rewards/chosen": -0.008947738446295261, | |
| "rewards/margins": 0.07473009079694748, | |
| "rewards/rejected": -0.08367783576250076, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.1020177413231334e-06, | |
| "logits/chosen": -0.2652960419654846, | |
| "logits/rejected": -0.25931379199028015, | |
| "logps/chosen": -435.1573791503906, | |
| "logps/rejected": -473.2171325683594, | |
| "loss": 0.1658, | |
| "rewards/accuracies": 0.5333333015441895, | |
| "rewards/chosen": -0.011026141233742237, | |
| "rewards/margins": 0.06725867092609406, | |
| "rewards/rejected": -0.07828481495380402, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.0446925746067768e-06, | |
| "logits/chosen": -0.28469547629356384, | |
| "logits/rejected": -0.21729762852191925, | |
| "logps/chosen": -415.32012939453125, | |
| "logps/rejected": -454.0088806152344, | |
| "loss": 0.1959, | |
| "rewards/accuracies": 0.49166664481163025, | |
| "rewards/chosen": -0.01757933758199215, | |
| "rewards/margins": 0.04614214226603508, | |
| "rewards/rejected": -0.06372148543596268, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.88502212844063e-07, | |
| "logits/chosen": -0.18305090069770813, | |
| "logits/rejected": -0.28139248490333557, | |
| "logps/chosen": -462.94598388671875, | |
| "logps/rejected": -518.9043579101562, | |
| "loss": 0.1791, | |
| "rewards/accuracies": 0.5416666865348816, | |
| "rewards/chosen": -0.022413870319724083, | |
| "rewards/margins": 0.06975697726011276, | |
| "rewards/rejected": -0.092170849442482, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 9.334904715888496e-07, | |
| "logits/chosen": -0.2571960687637329, | |
| "logits/rejected": -0.1717800498008728, | |
| "logps/chosen": -448.0375061035156, | |
| "logps/rejected": -492.38494873046875, | |
| "loss": 0.1906, | |
| "rewards/accuracies": 0.48333340883255005, | |
| "rewards/chosen": -0.01861201599240303, | |
| "rewards/margins": 0.05696520954370499, | |
| "rewards/rejected": -0.07557722926139832, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 8.797002473421729e-07, | |
| "logits/chosen": -0.24043762683868408, | |
| "logits/rejected": -0.22050480544567108, | |
| "logps/chosen": -460.51629638671875, | |
| "logps/rejected": -494.3642578125, | |
| "loss": 0.1907, | |
| "rewards/accuracies": 0.5333333611488342, | |
| "rewards/chosen": -0.024757737293839455, | |
| "rewards/margins": 0.05093444511294365, | |
| "rewards/rejected": -0.07569219172000885, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.271734841028553e-07, | |
| "logits/chosen": -0.2750247120857239, | |
| "logits/rejected": -0.1990630328655243, | |
| "logps/chosen": -414.51177978515625, | |
| "logps/rejected": -439.66583251953125, | |
| "loss": 0.1875, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -0.030453836545348167, | |
| "rewards/margins": 0.041512638330459595, | |
| "rewards/rejected": -0.07196646928787231, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.759511406608255e-07, | |
| "logits/chosen": -0.10888031870126724, | |
| "logits/rejected": -0.2525996267795563, | |
| "logps/chosen": -417.1622009277344, | |
| "logps/rejected": -480.26812744140625, | |
| "loss": 0.1814, | |
| "rewards/accuracies": 0.5291666984558105, | |
| "rewards/chosen": -0.019586481153964996, | |
| "rewards/margins": 0.0609549880027771, | |
| "rewards/rejected": -0.08054147660732269, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 7.260731586586983e-07, | |
| "logits/chosen": -0.28881603479385376, | |
| "logits/rejected": -0.22749415040016174, | |
| "logps/chosen": -418.9693298339844, | |
| "logps/rejected": -488.30487060546875, | |
| "loss": 0.1695, | |
| "rewards/accuracies": 0.5458333492279053, | |
| "rewards/chosen": -0.025732625275850296, | |
| "rewards/margins": 0.06428026407957077, | |
| "rewards/rejected": -0.09001289308071136, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 6.775784314464717e-07, | |
| "logits/chosen": -0.2233542948961258, | |
| "logits/rejected": -0.21398404240608215, | |
| "logps/chosen": -427.0870056152344, | |
| "logps/rejected": -473.38299560546875, | |
| "loss": 0.1746, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.015530099160969257, | |
| "rewards/margins": 0.059838373214006424, | |
| "rewards/rejected": -0.0753684714436531, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 6.305047737536707e-07, | |
| "logits/chosen": -0.2794179916381836, | |
| "logits/rejected": -0.1606103777885437, | |
| "logps/chosen": -443.2103576660156, | |
| "logps/rejected": -475.12835693359375, | |
| "loss": 0.1745, | |
| "rewards/accuracies": 0.5083333253860474, | |
| "rewards/chosen": -0.02874937281012535, | |
| "rewards/margins": 0.04828978329896927, | |
| "rewards/rejected": -0.07703914493322372, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 5.848888922025553e-07, | |
| "logits/chosen": -0.2493915855884552, | |
| "logits/rejected": -0.24869143962860107, | |
| "logps/chosen": -401.45263671875, | |
| "logps/rejected": -445.24774169921875, | |
| "loss": 0.1821, | |
| "rewards/accuracies": 0.4791666865348816, | |
| "rewards/chosen": -0.0232832133769989, | |
| "rewards/margins": 0.05199895054101944, | |
| "rewards/rejected": -0.07528215646743774, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5.407663566854008e-07, | |
| "logits/chosen": -0.2700728476047516, | |
| "logits/rejected": -0.2861558794975281, | |
| "logps/chosen": -420.72088623046875, | |
| "logps/rejected": -501.34991455078125, | |
| "loss": 0.1614, | |
| "rewards/accuracies": 0.5541666746139526, | |
| "rewards/chosen": -0.029055217280983925, | |
| "rewards/margins": 0.06994070112705231, | |
| "rewards/rejected": -0.09899591654539108, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.981715726281666e-07, | |
| "logits/chosen": -0.17120757699012756, | |
| "logits/rejected": -0.2849826216697693, | |
| "logps/chosen": -431.4073181152344, | |
| "logps/rejected": -485.0538024902344, | |
| "loss": 0.1867, | |
| "rewards/accuracies": 0.5208333730697632, | |
| "rewards/chosen": -0.028083154931664467, | |
| "rewards/margins": 0.050588030368089676, | |
| "rewards/rejected": -0.0786711797118187, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 4.5713775416217884e-07, | |
| "logits/chosen": -0.3271382451057434, | |
| "logits/rejected": -0.24611802399158478, | |
| "logps/chosen": -447.4208068847656, | |
| "logps/rejected": -512.2904052734375, | |
| "loss": 0.1849, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.02718830667436123, | |
| "rewards/margins": 0.05544273182749748, | |
| "rewards/rejected": -0.08263102918863297, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 4.1769689822475147e-07, | |
| "logits/chosen": -0.24327817559242249, | |
| "logits/rejected": -0.24398300051689148, | |
| "logps/chosen": -418.857177734375, | |
| "logps/rejected": -455.48052978515625, | |
| "loss": 0.1898, | |
| "rewards/accuracies": 0.44999998807907104, | |
| "rewards/chosen": -0.026202013716101646, | |
| "rewards/margins": 0.04488568380475044, | |
| "rewards/rejected": -0.07108769565820694, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 3.798797596089351e-07, | |
| "logits/chosen": -0.22220292687416077, | |
| "logits/rejected": -0.1325482428073883, | |
| "logps/chosen": -440.03704833984375, | |
| "logps/rejected": -495.2511291503906, | |
| "loss": 0.1807, | |
| "rewards/accuracies": 0.5041666626930237, | |
| "rewards/chosen": -0.028768355026841164, | |
| "rewards/margins": 0.05239064618945122, | |
| "rewards/rejected": -0.08115899562835693, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.4371582698185636e-07, | |
| "logits/chosen": -0.2665029764175415, | |
| "logits/rejected": -0.26157745718955994, | |
| "logps/chosen": -386.71441650390625, | |
| "logps/rejected": -432.8797302246094, | |
| "loss": 0.1877, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.021984888240695, | |
| "rewards/margins": 0.048461947590112686, | |
| "rewards/rejected": -0.07044683396816254, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 3.092332998903416e-07, | |
| "logits/chosen": -0.25679802894592285, | |
| "logits/rejected": -0.22036735713481903, | |
| "logps/chosen": -460.5780334472656, | |
| "logps/rejected": -521.6238403320312, | |
| "loss": 0.177, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.020292013883590698, | |
| "rewards/margins": 0.06953073292970657, | |
| "rewards/rejected": -0.08982274681329727, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 2.764590667717562e-07, | |
| "logits/chosen": -0.20977798104286194, | |
| "logits/rejected": -0.25232791900634766, | |
| "logps/chosen": -441.75091552734375, | |
| "logps/rejected": -503.1982421875, | |
| "loss": 0.1922, | |
| "rewards/accuracies": 0.5041667222976685, | |
| "rewards/chosen": -0.026740949600934982, | |
| "rewards/margins": 0.05636667460203171, | |
| "rewards/rejected": -0.08310762792825699, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.454186839872158e-07, | |
| "logits/chosen": -0.22725868225097656, | |
| "logits/rejected": -0.2418319284915924, | |
| "logps/chosen": -385.9988098144531, | |
| "logps/rejected": -447.5536193847656, | |
| "loss": 0.1749, | |
| "rewards/accuracies": 0.4583333134651184, | |
| "rewards/chosen": -0.014233797788619995, | |
| "rewards/margins": 0.05856531113386154, | |
| "rewards/rejected": -0.07279910892248154, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 2.1613635589349756e-07, | |
| "logits/chosen": -0.23666933178901672, | |
| "logits/rejected": -0.2799440324306488, | |
| "logps/chosen": -413.6375427246094, | |
| "logps/rejected": -459.60101318359375, | |
| "loss": 0.1858, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.019571740180253983, | |
| "rewards/margins": 0.051161814481019974, | |
| "rewards/rejected": -0.07073356211185455, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.8863491596921745e-07, | |
| "logits/chosen": -0.25514012575149536, | |
| "logits/rejected": -0.26672855019569397, | |
| "logps/chosen": -429.20318603515625, | |
| "logps/rejected": -485.3815002441406, | |
| "loss": 0.1865, | |
| "rewards/accuracies": 0.5083333253860474, | |
| "rewards/chosen": -0.02379278652369976, | |
| "rewards/margins": 0.0564563162624836, | |
| "rewards/rejected": -0.08024909347295761, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.629358090099639e-07, | |
| "logits/chosen": -0.2413201630115509, | |
| "logits/rejected": -0.30872708559036255, | |
| "logps/chosen": -447.03985595703125, | |
| "logps/rejected": -493.2216796875, | |
| "loss": 0.189, | |
| "rewards/accuracies": 0.4791666567325592, | |
| "rewards/chosen": -0.030030418187379837, | |
| "rewards/margins": 0.05704699829220772, | |
| "rewards/rejected": -0.08707742393016815, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.3905907440629752e-07, | |
| "logits/chosen": -0.23792843520641327, | |
| "logits/rejected": -0.28303149342536926, | |
| "logps/chosen": -416.78448486328125, | |
| "logps/rejected": -434.672119140625, | |
| "loss": 0.1939, | |
| "rewards/accuracies": 0.4791666865348816, | |
| "rewards/chosen": -0.018583226948976517, | |
| "rewards/margins": 0.041258666664361954, | |
| "rewards/rejected": -0.05984188988804817, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.1702333051763271e-07, | |
| "logits/chosen": -0.22025823593139648, | |
| "logits/rejected": -0.23834295570850372, | |
| "logps/chosen": -389.4822692871094, | |
| "logps/rejected": -454.0912170410156, | |
| "loss": 0.1826, | |
| "rewards/accuracies": 0.4791666865348816, | |
| "rewards/chosen": -0.019084876403212547, | |
| "rewards/margins": 0.052420567721128464, | |
| "rewards/rejected": -0.07150544971227646, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 9.684576015420277e-08, | |
| "logits/chosen": -0.2792247235774994, | |
| "logits/rejected": -0.1931079924106598, | |
| "logps/chosen": -450.61297607421875, | |
| "logps/rejected": -475.45947265625, | |
| "loss": 0.1857, | |
| "rewards/accuracies": 0.46666669845581055, | |
| "rewards/chosen": -0.018364811316132545, | |
| "rewards/margins": 0.049478866159915924, | |
| "rewards/rejected": -0.06784368306398392, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 7.854209717842231e-08, | |
| "logits/chosen": -0.2361956536769867, | |
| "logits/rejected": -0.16988393664360046, | |
| "logps/chosen": -463.72625732421875, | |
| "logps/rejected": -532.1114501953125, | |
| "loss": 0.1735, | |
| "rewards/accuracies": 0.5750000476837158, | |
| "rewards/chosen": -0.01867097057402134, | |
| "rewards/margins": 0.07356119900941849, | |
| "rewards/rejected": -0.09223215281963348, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 6.212661423609184e-08, | |
| "logits/chosen": -0.2650465965270996, | |
| "logits/rejected": -0.22043271362781525, | |
| "logps/chosen": -476.1746520996094, | |
| "logps/rejected": -500.8211364746094, | |
| "loss": 0.1848, | |
| "rewards/accuracies": 0.491666704416275, | |
| "rewards/chosen": -0.02679665945470333, | |
| "rewards/margins": 0.05153702571988106, | |
| "rewards/rejected": -0.07833368331193924, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 4.761211162702117e-08, | |
| "logits/chosen": -0.2697955369949341, | |
| "logits/rejected": -0.2639797329902649, | |
| "logps/chosen": -401.6107177734375, | |
| "logps/rejected": -457.95703125, | |
| "loss": 0.1775, | |
| "rewards/accuracies": 0.4833333492279053, | |
| "rewards/chosen": -0.02222241833806038, | |
| "rewards/margins": 0.05034973472356796, | |
| "rewards/rejected": -0.07257215678691864, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 3.5009907323737826e-08, | |
| "logits/chosen": -0.2754846215248108, | |
| "logits/rejected": -0.3097229599952698, | |
| "logps/chosen": -415.70867919921875, | |
| "logps/rejected": -478.73663330078125, | |
| "loss": 0.1718, | |
| "rewards/accuracies": 0.5291666984558105, | |
| "rewards/chosen": -0.013669279403984547, | |
| "rewards/margins": 0.07036744058132172, | |
| "rewards/rejected": -0.08403672277927399, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 2.4329828146074096e-08, | |
| "logits/chosen": -0.2891579866409302, | |
| "logits/rejected": -0.19230952858924866, | |
| "logps/chosen": -454.9552307128906, | |
| "logps/rejected": -495.06488037109375, | |
| "loss": 0.1764, | |
| "rewards/accuracies": 0.5708333849906921, | |
| "rewards/chosen": -0.016126099973917007, | |
| "rewards/margins": 0.06828036159276962, | |
| "rewards/rejected": -0.08440648019313812, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.5580202098509078e-08, | |
| "logits/chosen": -0.29824376106262207, | |
| "logits/rejected": -0.22227320075035095, | |
| "logps/chosen": -399.3383483886719, | |
| "logps/rejected": -465.268310546875, | |
| "loss": 0.1686, | |
| "rewards/accuracies": 0.5041666626930237, | |
| "rewards/chosen": -0.010598058812320232, | |
| "rewards/margins": 0.06828632950782776, | |
| "rewards/rejected": -0.07888438552618027, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 8.767851876239075e-09, | |
| "logits/chosen": -0.25601503252983093, | |
| "logits/rejected": -0.28030115365982056, | |
| "logps/chosen": -452.5179138183594, | |
| "logps/rejected": -499.35125732421875, | |
| "loss": 0.1759, | |
| "rewards/accuracies": 0.5500000715255737, | |
| "rewards/chosen": -0.017482977360486984, | |
| "rewards/margins": 0.05968954414129257, | |
| "rewards/rejected": -0.07717252522706985, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 3.8980895450474455e-09, | |
| "logits/chosen": -0.24264974892139435, | |
| "logits/rejected": -0.2742842137813568, | |
| "logps/chosen": -439.875244140625, | |
| "logps/rejected": -479.13873291015625, | |
| "loss": 0.1792, | |
| "rewards/accuracies": 0.5333333015441895, | |
| "rewards/chosen": -0.020408231765031815, | |
| "rewards/margins": 0.05712126940488815, | |
| "rewards/rejected": -0.07752950489521027, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 9.747123991141193e-10, | |
| "logits/chosen": -0.25368720293045044, | |
| "logits/rejected": -0.23606958985328674, | |
| "logps/chosen": -377.1168212890625, | |
| "logps/rejected": -435.25732421875, | |
| "loss": 0.1724, | |
| "rewards/accuracies": 0.4916667342185974, | |
| "rewards/chosen": -0.018287640064954758, | |
| "rewards/margins": 0.055953748524188995, | |
| "rewards/rejected": -0.07424138486385345, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0, | |
| "logits/chosen": -0.2323763072490692, | |
| "logits/rejected": -0.3323909640312195, | |
| "logps/chosen": -420.1419372558594, | |
| "logps/rejected": -468.536376953125, | |
| "loss": 0.1831, | |
| "rewards/accuracies": 0.4541666507720947, | |
| "rewards/chosen": -0.01915622688829899, | |
| "rewards/margins": 0.05424219369888306, | |
| "rewards/rejected": -0.07339842617511749, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 1250, | |
| "total_flos": 0.0, | |
| "train_loss": 0.1874461171746254, | |
| "train_runtime": 12039.6342, | |
| "train_samples_per_second": 2.492, | |
| "train_steps_per_second": 0.104 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 0.0, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |