{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.999247554552295,
  "eval_steps": 30,
  "global_step": 166,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.012039127163280662,
      "grad_norm": 11.953282356262207,
      "learning_rate": 5.555555555555555e-08,
      "logits/chosen": -0.48816660046577454,
      "logits/rejected": -0.42142170667648315,
      "logps/chosen": -117.26611328125,
      "logps/rejected": -125.41987609863281,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06019563581640331,
      "grad_norm": 16.68506622314453,
      "learning_rate": 2.7777777777777776e-07,
      "logits/chosen": -0.46595269441604614,
      "logits/rejected": -0.356529176235199,
      "logps/chosen": -190.95057678222656,
      "logps/rejected": -211.25076293945312,
      "loss": 0.6926,
      "rewards/accuracies": 0.453125,
      "rewards/chosen": 0.0007588082225993276,
      "rewards/margins": 0.0022044419310986996,
      "rewards/rejected": -0.0014456338249146938,
      "step": 5
    },
    {
      "epoch": 0.12039127163280662,
      "grad_norm": 13.722668647766113,
      "learning_rate": 4.999499509357132e-07,
      "logits/chosen": -0.4793759286403656,
      "logits/rejected": -0.37052756547927856,
      "logps/chosen": -155.6678009033203,
      "logps/rejected": -199.44947814941406,
      "loss": 0.6889,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": 0.005186144262552261,
      "rewards/margins": 0.009383995085954666,
      "rewards/rejected": -0.004197851754724979,
      "step": 10
    },
    {
      "epoch": 0.18058690744920994,
      "grad_norm": 12.493337631225586,
      "learning_rate": 4.982003369106287e-07,
      "logits/chosen": -0.49185729026794434,
      "logits/rejected": -0.37670475244522095,
      "logps/chosen": -76.30625915527344,
      "logps/rejected": -177.40817260742188,
      "loss": 0.6691,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.028129320591688156,
      "rewards/margins": 0.048517487943172455,
      "rewards/rejected": -0.020388163626194,
      "step": 15
    },
    {
      "epoch": 0.24078254326561324,
      "grad_norm": 11.79749870300293,
      "learning_rate": 4.939682729058838e-07,
      "logits/chosen": -0.45126277208328247,
      "logits/rejected": -0.3729521930217743,
      "logps/chosen": -166.20733642578125,
      "logps/rejected": -207.5035400390625,
      "loss": 0.6249,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.08626963198184967,
      "rewards/margins": 0.14315639436244965,
      "rewards/rejected": -0.05688678100705147,
      "step": 20
    },
    {
      "epoch": 0.3009781790820166,
      "grad_norm": 10.574383735656738,
      "learning_rate": 4.872960871766826e-07,
      "logits/chosen": -0.4710594713687897,
      "logits/rejected": -0.3694532513618469,
      "logps/chosen": -86.93299865722656,
      "logps/rejected": -186.15248107910156,
      "loss": 0.587,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.12460210174322128,
      "rewards/margins": 0.23137669265270233,
      "rewards/rejected": -0.10677458345890045,
      "step": 25
    },
    {
      "epoch": 0.3611738148984199,
      "grad_norm": 11.093204498291016,
      "learning_rate": 4.782505135862175e-07,
      "logits/chosen": -0.45945605635643005,
      "logits/rejected": -0.33354875445365906,
      "logps/chosen": -71.20188903808594,
      "logps/rejected": -212.95767211914062,
      "loss": 0.5229,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.16831260919570923,
      "rewards/margins": 0.38181251287460327,
      "rewards/rejected": -0.21349990367889404,
      "step": 30
    },
    {
      "epoch": 0.3611738148984199,
      "eval_logits/chosen": -0.4618959426879883,
      "eval_logits/rejected": -0.3433874249458313,
      "eval_logps/chosen": -98.28858947753906,
      "eval_logps/rejected": -212.0100555419922,
      "eval_loss": 0.5059286952018738,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 0.18195854127407074,
      "eval_rewards/margins": 0.4181654751300812,
      "eval_rewards/rejected": -0.23620688915252686,
      "eval_runtime": 179.4182,
      "eval_samples_per_second": 3.043,
      "eval_steps_per_second": 1.522,
      "step": 30
    },
    {
      "epoch": 0.4213694507148232,
      "grad_norm": 9.376220703125,
      "learning_rate": 4.6692202414695724e-07,
      "logits/chosen": -0.4632042944431305,
      "logits/rejected": -0.35029542446136475,
      "logps/chosen": -84.06396484375,
      "logps/rejected": -213.848388671875,
      "loss": 0.4976,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2034817487001419,
      "rewards/margins": 0.4632874131202698,
      "rewards/rejected": -0.25980567932128906,
      "step": 35
    },
    {
      "epoch": 0.4815650865312265,
      "grad_norm": 8.679346084594727,
      "learning_rate": 4.534239241377266e-07,
      "logits/chosen": -0.44362330436706543,
      "logits/rejected": -0.2992916703224182,
      "logps/chosen": -105.2283706665039,
      "logps/rejected": -244.84890747070312,
      "loss": 0.4197,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.22778573632240295,
      "rewards/margins": 0.6910415291786194,
      "rewards/rejected": -0.46325573325157166,
      "step": 40
    },
    {
      "epoch": 0.5417607223476298,
      "grad_norm": 7.219143867492676,
      "learning_rate": 4.3789121884703727e-07,
      "logits/chosen": -0.41270333528518677,
      "logits/rejected": -0.27924439311027527,
      "logps/chosen": -70.08865356445312,
      "logps/rejected": -261.56170654296875,
      "loss": 0.3621,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.23598209023475647,
      "rewards/margins": 0.9187321662902832,
      "rewards/rejected": -0.6827500462532043,
      "step": 45
    },
    {
      "epoch": 0.6019563581640331,
      "grad_norm": 6.640863418579102,
      "learning_rate": 4.204792632772754e-07,
      "logits/chosen": -0.4174782633781433,
      "logits/rejected": -0.2659801244735718,
      "logps/chosen": -109.1211166381836,
      "logps/rejected": -280.77813720703125,
      "loss": 0.3123,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2913265824317932,
      "rewards/margins": 1.1760694980621338,
      "rewards/rejected": -0.8847430348396301,
      "step": 50
    },
    {
      "epoch": 0.6621519939804364,
      "grad_norm": 5.293730735778809,
      "learning_rate": 4.01362208315132e-07,
      "logits/chosen": -0.4078051447868347,
      "logits/rejected": -0.25378990173339844,
      "logps/chosen": -116.1395492553711,
      "logps/rejected": -301.702392578125,
      "loss": 0.2619,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.3083065152168274,
      "rewards/margins": 1.4346027374267578,
      "rewards/rejected": -1.1262962818145752,
      "step": 55
    },
    {
      "epoch": 0.7223476297968398,
      "grad_norm": 4.923187255859375,
      "learning_rate": 3.807312589093701e-07,
      "logits/chosen": -0.4022981524467468,
      "logits/rejected": -0.2537968158721924,
      "logps/chosen": -103.5102310180664,
      "logps/rejected": -326.17486572265625,
      "loss": 0.2411,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2954918146133423,
      "rewards/margins": 1.6640812158584595,
      "rewards/rejected": -1.3685895204544067,
      "step": 60
    },
    {
      "epoch": 0.7223476297968398,
      "eval_logits/chosen": -0.406698077917099,
      "eval_logits/rejected": -0.23272451758384705,
      "eval_logps/chosen": -88.900146484375,
      "eval_logps/rejected": -330.7860107421875,
      "eval_loss": 0.2134791761636734,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 0.27584296464920044,
      "eval_rewards/margins": 1.699809193611145,
      "eval_rewards/rejected": -1.4239662885665894,
      "eval_runtime": 183.6706,
      "eval_samples_per_second": 2.973,
      "eval_steps_per_second": 1.486,
      "step": 60
    },
    {
      "epoch": 0.782543265613243,
      "grad_norm": 4.418694496154785,
      "learning_rate": 3.5879276167728337e-07,
      "logits/chosen": -0.4011690616607666,
      "logits/rejected": -0.22693100571632385,
      "logps/chosen": -56.017845153808594,
      "logps/rejected": -332.90380859375,
      "loss": 0.1992,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2577177882194519,
      "rewards/margins": 1.7556695938110352,
      "rewards/rejected": -1.497951865196228,
      "step": 65
    },
    {
      "epoch": 0.8427389014296464,
      "grad_norm": 3.794067859649658,
      "learning_rate": 3.357661410672247e-07,
      "logits/chosen": -0.33221831917762756,
      "logits/rejected": -0.1342475712299347,
      "logps/chosen": -74.8525619506836,
      "logps/rejected": -393.6372985839844,
      "loss": 0.1573,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.25318774580955505,
      "rewards/margins": 2.2917141914367676,
      "rewards/rejected": -2.0385265350341797,
      "step": 70
    },
    {
      "epoch": 0.9029345372460497,
      "grad_norm": 3.2060582637786865,
      "learning_rate": 3.1188170471929064e-07,
      "logits/chosen": -0.2731170058250427,
      "logits/rejected": -0.10557065159082413,
      "logps/chosen": -161.33474731445312,
      "logps/rejected": -437.1578063964844,
      "loss": 0.1191,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2284388542175293,
      "rewards/margins": 2.7684743404388428,
      "rewards/rejected": -2.5400352478027344,
      "step": 75
    },
    {
      "epoch": 0.963130173062453,
      "grad_norm": 1.8243048191070557,
      "learning_rate": 2.8737833997450657e-07,
      "logits/chosen": -0.2729615569114685,
      "logits/rejected": -0.0838087797164917,
      "logps/chosen": -80.7784423828125,
      "logps/rejected": -492.26080322265625,
      "loss": 0.0926,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.21951308846473694,
      "rewards/margins": 3.2957847118377686,
      "rewards/rejected": -3.0762715339660645,
      "step": 80
    },
    {
      "epoch": 1.0233258088788564,
      "grad_norm": 1.6663548946380615,
      "learning_rate": 2.6250112457156293e-07,
      "logits/chosen": -0.2614014744758606,
      "logits/rejected": -0.06510574370622635,
      "logps/chosen": -87.82209777832031,
      "logps/rejected": -556.6070556640625,
      "loss": 0.0775,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.1841917783021927,
      "rewards/margins": 3.8404979705810547,
      "rewards/rejected": -3.656306028366089,
      "step": 85
    },
    {
      "epoch": 1.0835214446952597,
      "grad_norm": 1.4261465072631836,
      "learning_rate": 2.3749887542843707e-07,
      "logits/chosen": -0.26909708976745605,
      "logits/rejected": -0.0703195109963417,
      "logps/chosen": -100.4935531616211,
      "logps/rejected": -598.0264892578125,
      "loss": 0.0634,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.16738824546337128,
      "rewards/margins": 4.255741119384766,
      "rewards/rejected": -4.088352680206299,
      "step": 90
    },
    {
      "epoch": 1.0835214446952597,
      "eval_logits/chosen": -0.25795042514801025,
      "eval_logits/rejected": -0.035701148211956024,
      "eval_logps/chosen": -99.51206970214844,
      "eval_logps/rejected": -607.3591918945312,
      "eval_loss": 0.07514728605747223,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 0.16972379386425018,
      "eval_rewards/margins": 4.359421730041504,
      "eval_rewards/rejected": -4.189698219299316,
      "eval_runtime": 184.8502,
      "eval_samples_per_second": 2.954,
      "eval_steps_per_second": 1.477,
      "step": 90
    },
    {
      "epoch": 1.143717080511663,
      "grad_norm": 1.3145774602890015,
      "learning_rate": 2.126216600254934e-07,
      "logits/chosen": -0.2394520789384842,
      "logits/rejected": -0.023534994572401047,
      "logps/chosen": -150.5535888671875,
      "logps/rejected": -699.5059814453125,
      "loss": 0.0605,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.09343000501394272,
      "rewards/margins": 5.061221122741699,
      "rewards/rejected": -4.967791557312012,
      "step": 95
    },
    {
      "epoch": 1.2039127163280663,
      "grad_norm": 0.628934919834137,
      "learning_rate": 1.8811829528070931e-07,
      "logits/chosen": -0.2859761714935303,
      "logits/rejected": -0.027079975232481956,
      "logps/chosen": -72.18778991699219,
      "logps/rejected": -737.0675048828125,
      "loss": 0.0489,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.1680155098438263,
      "rewards/margins": 5.535449028015137,
      "rewards/rejected": -5.367433547973633,
      "step": 100
    },
    {
      "epoch": 1.2648607975921746,
      "grad_norm": 0.8070765733718872,
      "learning_rate": 1.6423385893277537e-07,
      "logits/chosen": -0.24801869690418243,
      "logits/rejected": -0.014291681349277496,
      "logps/chosen": -109.48841857910156,
      "logps/rejected": -681.1419677734375,
      "loss": 0.0476,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.09798868745565414,
      "rewards/margins": 5.150273323059082,
      "rewards/rejected": -5.052285194396973,
      "step": 105
    },
    {
      "epoch": 1.325056433408578,
      "grad_norm": 0.6888783574104309,
      "learning_rate": 1.4120723832271663e-07,
      "logits/chosen": -0.23577141761779785,
      "logits/rejected": -0.011897795833647251,
      "logps/chosen": -120.05766296386719,
      "logps/rejected": -731.9443969726562,
      "loss": 0.054,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.07691726088523865,
      "rewards/margins": 5.525856018066406,
      "rewards/rejected": -5.448939323425293,
      "step": 110
    },
    {
      "epoch": 1.3852520692249812,
      "grad_norm": 0.6200137734413147,
      "learning_rate": 1.1926874109062998e-07,
      "logits/chosen": -0.2343941181898117,
      "logits/rejected": 0.012432652525603771,
      "logps/chosen": -131.77175903320312,
      "logps/rejected": -745.412841796875,
      "loss": 0.0496,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.02012884058058262,
      "rewards/margins": 5.680893898010254,
      "rewards/rejected": -5.660765171051025,
      "step": 115
    },
    {
      "epoch": 1.4454477050413845,
      "grad_norm": 0.5668838620185852,
      "learning_rate": 9.863779168486797e-08,
      "logits/chosen": -0.22034311294555664,
      "logits/rejected": 0.03210270777344704,
      "logps/chosen": -115.6126708984375,
      "logps/rejected": -790.9016723632812,
      "loss": 0.0452,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.053596943616867065,
      "rewards/margins": 6.07260799407959,
      "rewards/rejected": -6.0190110206604,
      "step": 120
    },
    {
      "epoch": 1.4454477050413845,
      "eval_logits/chosen": -0.2345404177904129,
      "eval_logits/rejected": 0.038044609129428864,
      "eval_logps/chosen": -108.91590881347656,
      "eval_logps/rejected": -782.349365234375,
      "eval_loss": 0.05323062837123871,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 0.07568521797657013,
      "eval_rewards/margins": 6.015285491943359,
      "eval_rewards/rejected": -5.939600467681885,
      "eval_runtime": 191.5149,
      "eval_samples_per_second": 2.851,
      "eval_steps_per_second": 1.425,
      "step": 120
    },
    {
      "epoch": 1.5056433408577878,
      "grad_norm": 0.7317198514938354,
      "learning_rate": 7.952073672272464e-08,
      "logits/chosen": -0.21637864410877228,
      "logits/rejected": 0.0313236340880394,
      "logps/chosen": -126.0725326538086,
      "logps/rejected": -756.0071411132812,
      "loss": 0.0492,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.008594411425292492,
      "rewards/margins": 5.820641994476318,
      "rewards/rejected": -5.82923698425293,
      "step": 125
    },
    {
      "epoch": 1.5658389766741911,
      "grad_norm": 0.5411990284919739,
      "learning_rate": 6.210878115296267e-08,
      "logits/chosen": -0.2258405238389969,
      "logits/rejected": 0.026202013716101646,
      "logps/chosen": -144.01856994628906,
      "logps/rejected": -760.4769287109375,
      "loss": 0.029,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.01425357349216938,
      "rewards/margins": 5.899747371673584,
      "rewards/rejected": -5.91400146484375,
      "step": 130
    },
    {
      "epoch": 1.6260346124905944,
      "grad_norm": 0.30838528275489807,
      "learning_rate": 4.657607586227344e-08,
      "logits/chosen": -0.23345847427845,
      "logits/rejected": 0.027966421097517014,
      "logps/chosen": -124.6040267944336,
      "logps/rejected": -917.08349609375,
      "loss": 0.036,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.0217633955180645,
      "rewards/margins": 7.151785850524902,
      "rewards/rejected": -7.1300225257873535,
      "step": 135
    },
    {
      "epoch": 1.6862302483069977,
      "grad_norm": 1.47968590259552,
      "learning_rate": 3.30779758530427e-08,
      "logits/chosen": -0.2245132029056549,
      "logits/rejected": 0.047028228640556335,
      "logps/chosen": -118.25953674316406,
      "logps/rejected": -851.4835205078125,
      "loss": 0.0516,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -0.0352528840303421,
      "rewards/margins": 6.62514591217041,
      "rewards/rejected": -6.660399436950684,
      "step": 140
    },
    {
      "epoch": 1.746425884123401,
      "grad_norm": 0.5548922419548035,
      "learning_rate": 2.1749486413782435e-08,
      "logits/chosen": -0.21600095927715302,
      "logits/rejected": 0.06509985029697418,
      "logps/chosen": -140.62574768066406,
      "logps/rejected": -936.2386474609375,
      "loss": 0.0312,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.07837997376918793,
      "rewards/margins": 7.2990617752075195,
      "rewards/rejected": -7.37744140625,
      "step": 145
    },
    {
      "epoch": 1.8066215199398044,
      "grad_norm": 0.5003569722175598,
      "learning_rate": 1.2703912823317397e-08,
      "logits/chosen": -0.20677892863750458,
      "logits/rejected": 0.06903555244207382,
      "logps/chosen": -150.6492919921875,
      "logps/rejected": -826.0875244140625,
      "loss": 0.0307,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.09083503484725952,
      "rewards/margins": 6.384757995605469,
      "rewards/rejected": -6.475593566894531,
      "step": 150
    },
    {
      "epoch": 1.8066215199398044,
      "eval_logits/chosen": -0.230697363615036,
      "eval_logits/rejected": 0.079569511115551,
      "eval_logps/chosen": -114.66878509521484,
      "eval_logps/rejected": -867.22607421875,
      "eval_loss": 0.045937325805425644,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": 0.018156491219997406,
      "eval_rewards/margins": 6.806523323059082,
      "eval_rewards/rejected": -6.78836727142334,
      "eval_runtime": 193.7203,
      "eval_samples_per_second": 2.818,
      "eval_steps_per_second": 1.409,
      "step": 150
    },
    {
      "epoch": 1.8668171557562077,
      "grad_norm": 0.303206205368042,
      "learning_rate": 6.031727094116174e-09,
      "logits/chosen": -0.22335031628608704,
      "logits/rejected": 0.0749906450510025,
      "logps/chosen": -178.22097778320312,
      "logps/rejected": -1008.4031372070312,
      "loss": 0.0311,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.15972693264484406,
      "rewards/margins": 7.850977420806885,
      "rewards/rejected": -8.01070499420166,
      "step": 155
    },
    {
      "epoch": 1.927012791572611,
      "grad_norm": 0.5809817910194397,
      "learning_rate": 1.7996630893712671e-09,
      "logits/chosen": -0.20071235299110413,
      "logits/rejected": 0.08598125725984573,
      "logps/chosen": -156.4674530029297,
      "logps/rejected": -796.1658325195312,
      "loss": 0.0451,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.11812801659107208,
      "rewards/margins": 6.20440673828125,
      "rewards/rejected": -6.322534084320068,
      "step": 160
    },
    {
      "epoch": 1.9872084273890143,
      "grad_norm": 0.5935037732124329,
      "learning_rate": 5.0049064286850074e-11,
      "logits/chosen": -0.20801086723804474,
      "logits/rejected": 0.09084095805883408,
      "logps/chosen": -133.05039978027344,
      "logps/rejected": -858.1156005859375,
      "loss": 0.0449,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.057532183825969696,
      "rewards/margins": 6.773335933685303,
      "rewards/rejected": -6.830867767333984,
      "step": 165
    },
    {
      "epoch": 1.999247554552295,
      "step": 166,
      "total_flos": 0.0,
      "train_loss": 0.01645123219139964,
      "train_runtime": 7014.6347,
      "train_samples_per_second": 0.758,
      "train_steps_per_second": 0.024
    }
  ],
  "logging_steps": 5,
  "max_steps": 166,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "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": 1,
  "trial_name": null,
  "trial_params": null
}