{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 50,
  "global_step": 201,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07462686567164178,
      "grad_norm": 53.89263578664345,
      "learning_rate": 5e-07,
      "logits/chosen": -2.747152805328369,
      "logits/rejected": -2.7397713661193848,
      "logps/chosen": -254.6358642578125,
      "logps/rejected": -205.4340057373047,
      "loss": 0.6908,
      "rewards/accuracies": 0.3375000059604645,
      "rewards/chosen": 0.01600971445441246,
      "rewards/margins": 0.003508391324430704,
      "rewards/rejected": 0.012501319870352745,
      "step": 5
    },
    {
      "epoch": 0.14925373134328357,
      "grad_norm": 51.3561964257765,
      "learning_rate": 1e-06,
      "logits/chosen": -2.701726198196411,
      "logits/rejected": -2.6838648319244385,
      "logps/chosen": -239.114501953125,
      "logps/rejected": -208.04202270507812,
      "loss": 0.6434,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": 0.47272032499313354,
      "rewards/margins": 0.1491888463497162,
      "rewards/rejected": 0.3235314190387726,
      "step": 10
    },
    {
      "epoch": 0.22388059701492538,
      "grad_norm": 47.770878647824766,
      "learning_rate": 9.983100718730718e-07,
      "logits/chosen": -2.455693006515503,
      "logits/rejected": -2.4368350505828857,
      "logps/chosen": -238.6224365234375,
      "logps/rejected": -190.65304565429688,
      "loss": 0.6652,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": 1.1667336225509644,
      "rewards/margins": 0.4549005925655365,
      "rewards/rejected": 0.711833119392395,
      "step": 15
    },
    {
      "epoch": 0.29850746268656714,
      "grad_norm": 41.77400166740322,
      "learning_rate": 9.932517109205849e-07,
      "logits/chosen": -2.3191988468170166,
      "logits/rejected": -2.298821210861206,
      "logps/chosen": -234.3042755126953,
      "logps/rejected": -199.0926971435547,
      "loss": 0.6162,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": 1.481550931930542,
      "rewards/margins": 0.7337996363639832,
      "rewards/rejected": 0.7477512955665588,
      "step": 20
    },
    {
      "epoch": 0.373134328358209,
      "grad_norm": 37.097180665689756,
      "learning_rate": 9.848591102083375e-07,
      "logits/chosen": -2.243551731109619,
      "logits/rejected": -2.2224888801574707,
      "logps/chosen": -270.208251953125,
      "logps/rejected": -196.4565887451172,
      "loss": 0.596,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": 1.8057639598846436,
      "rewards/margins": 1.2407209873199463,
      "rewards/rejected": 0.565043032169342,
      "step": 25
    },
    {
      "epoch": 0.44776119402985076,
      "grad_norm": 40.16458707164062,
      "learning_rate": 9.731890013043367e-07,
      "logits/chosen": -2.2443346977233887,
      "logits/rejected": -2.251404285430908,
      "logps/chosen": -231.3487548828125,
      "logps/rejected": -221.5731964111328,
      "loss": 0.5597,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 1.3293423652648926,
      "rewards/margins": 1.2652809619903564,
      "rewards/rejected": 0.0640614926815033,
      "step": 30
    },
    {
      "epoch": 0.5223880597014925,
      "grad_norm": 46.613500144353665,
      "learning_rate": 9.583202707897073e-07,
      "logits/chosen": -2.3597989082336426,
      "logits/rejected": -2.3549132347106934,
      "logps/chosen": -241.8134765625,
      "logps/rejected": -214.6074676513672,
      "loss": 0.6498,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": 0.8961586952209473,
      "rewards/margins": 0.7394484877586365,
      "rewards/rejected": 0.15671022236347198,
      "step": 35
    },
    {
      "epoch": 0.5970149253731343,
      "grad_norm": 42.571655553623124,
      "learning_rate": 9.403534270080829e-07,
      "logits/chosen": -2.3809351921081543,
      "logits/rejected": -2.3687262535095215,
      "logps/chosen": -239.2801513671875,
      "logps/rejected": -209.7950439453125,
      "loss": 0.6099,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": 1.0974634885787964,
      "rewards/margins": 1.2669336795806885,
      "rewards/rejected": -0.16947026550769806,
      "step": 40
    },
    {
      "epoch": 0.6716417910447762,
      "grad_norm": 40.02985298301046,
      "learning_rate": 9.19409920658098e-07,
      "logits/chosen": -2.2248692512512207,
      "logits/rejected": -2.2323927879333496,
      "logps/chosen": -262.9397888183594,
      "logps/rejected": -202.8191680908203,
      "loss": 0.612,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/chosen": 1.1908448934555054,
      "rewards/margins": 1.1182587146759033,
      "rewards/rejected": 0.07258612662553787,
      "step": 45
    },
    {
      "epoch": 0.746268656716418,
      "grad_norm": 44.22077560247569,
      "learning_rate": 8.956313238215823e-07,
      "logits/chosen": -2.1219964027404785,
      "logits/rejected": -2.0875980854034424,
      "logps/chosen": -242.64578247070312,
      "logps/rejected": -204.98623657226562,
      "loss": 0.6007,
      "rewards/accuracies": 0.768750011920929,
      "rewards/chosen": 1.1351102590560913,
      "rewards/margins": 1.0577830076217651,
      "rewards/rejected": 0.07732732594013214,
      "step": 50
    },
    {
      "epoch": 0.746268656716418,
      "eval_logits/chosen": -2.118994951248169,
      "eval_logits/rejected": -2.085069417953491,
      "eval_logps/chosen": -264.02423095703125,
      "eval_logps/rejected": -229.3638458251953,
      "eval_loss": 0.6151112914085388,
      "eval_rewards/accuracies": 0.762499988079071,
      "eval_rewards/chosen": 1.2504631280899048,
      "eval_rewards/margins": 1.4133154153823853,
      "eval_rewards/rejected": -0.16285225749015808,
      "eval_runtime": 128.217,
      "eval_samples_per_second": 14.819,
      "eval_steps_per_second": 0.234,
      "step": 50
    },
    {
      "epoch": 0.8208955223880597,
      "grad_norm": 40.18630443466436,
      "learning_rate": 8.691783729769873e-07,
      "logits/chosen": -2.11806321144104,
      "logits/rejected": -2.0859131813049316,
      "logps/chosen": -245.06289672851562,
      "logps/rejected": -208.0782928466797,
      "loss": 0.5871,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": 0.9679476618766785,
      "rewards/margins": 1.4604319334030151,
      "rewards/rejected": -0.49248433113098145,
      "step": 55
    },
    {
      "epoch": 0.8955223880597015,
      "grad_norm": 35.47562973498541,
      "learning_rate": 8.402298824670029e-07,
      "logits/chosen": -2.089561939239502,
      "logits/rejected": -2.082063674926758,
      "logps/chosen": -249.4997100830078,
      "logps/rejected": -218.75180053710938,
      "loss": 0.5522,
      "rewards/accuracies": 0.75,
      "rewards/chosen": 0.7627511024475098,
      "rewards/margins": 1.2879369258880615,
      "rewards/rejected": -0.5251858830451965,
      "step": 60
    },
    {
      "epoch": 0.9701492537313433,
      "grad_norm": 33.55304707235284,
      "learning_rate": 8.089815357650089e-07,
      "logits/chosen": -2.074007749557495,
      "logits/rejected": -2.046736717224121,
      "logps/chosen": -254.79116821289062,
      "logps/rejected": -216.1507110595703,
      "loss": 0.5687,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": 0.8360864520072937,
      "rewards/margins": 1.374267816543579,
      "rewards/rejected": -0.5381813049316406,
      "step": 65
    },
    {
      "epoch": 1.044776119402985,
      "grad_norm": 21.072890393731228,
      "learning_rate": 7.756445627110522e-07,
      "logits/chosen": -2.046535015106201,
      "logits/rejected": -2.0300545692443848,
      "logps/chosen": -242.8757781982422,
      "logps/rejected": -222.74545288085938,
      "loss": 0.4086,
      "rewards/accuracies": 0.831250011920929,
      "rewards/chosen": 1.4517204761505127,
      "rewards/margins": 2.1768176555633545,
      "rewards/rejected": -0.7250972986221313,
      "step": 70
    },
    {
      "epoch": 1.1194029850746268,
      "grad_norm": 22.42027466653792,
      "learning_rate": 7.404443116588547e-07,
      "logits/chosen": -2.0193662643432617,
      "logits/rejected": -1.9956552982330322,
      "logps/chosen": -241.31900024414062,
      "logps/rejected": -227.45822143554688,
      "loss": 0.2538,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": 1.9737942218780518,
      "rewards/margins": 2.759068489074707,
      "rewards/rejected": -0.7852741479873657,
      "step": 75
    },
    {
      "epoch": 1.1940298507462686,
      "grad_norm": 16.719506543900728,
      "learning_rate": 7.036187261857288e-07,
      "logits/chosen": -2.040085554122925,
      "logits/rejected": -1.9938533306121826,
      "logps/chosen": -238.9818572998047,
      "logps/rejected": -231.0389862060547,
      "loss": 0.2171,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": 2.3720481395721436,
      "rewards/margins": 3.3235080242156982,
      "rewards/rejected": -0.9514597654342651,
      "step": 80
    },
    {
      "epoch": 1.2686567164179103,
      "grad_norm": 26.325361102777258,
      "learning_rate": 6.654167366624008e-07,
      "logits/chosen": -2.0701422691345215,
      "logits/rejected": -2.0505211353302,
      "logps/chosen": -225.2556915283203,
      "logps/rejected": -230.9950408935547,
      "loss": 0.2527,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 2.3914668560028076,
      "rewards/margins": 3.2805893421173096,
      "rewards/rejected": -0.8891223669052124,
      "step": 85
    },
    {
      "epoch": 1.3432835820895521,
      "grad_norm": 22.27953814968023,
      "learning_rate": 6.260965775552713e-07,
      "logits/chosen": -2.1463778018951416,
      "logits/rejected": -2.1030120849609375,
      "logps/chosen": -232.5117950439453,
      "logps/rejected": -203.0765838623047,
      "loss": 0.2471,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": 2.4263741970062256,
      "rewards/margins": 3.221815586090088,
      "rewards/rejected": -0.7954414486885071,
      "step": 90
    },
    {
      "epoch": 1.417910447761194,
      "grad_norm": 25.58307192903391,
      "learning_rate": 5.859240418356614e-07,
      "logits/chosen": -2.1819968223571777,
      "logits/rejected": -2.145015239715576,
      "logps/chosen": -227.00015258789062,
      "logps/rejected": -219.8661346435547,
      "loss": 0.3243,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 2.5345447063446045,
      "rewards/margins": 3.464165449142456,
      "rewards/rejected": -0.9296209216117859,
      "step": 95
    },
    {
      "epoch": 1.4925373134328357,
      "grad_norm": 22.088806026665704,
      "learning_rate": 5.451706842957421e-07,
      "logits/chosen": -2.2039341926574707,
      "logits/rejected": -2.186410427093506,
      "logps/chosen": -221.75912475585938,
      "logps/rejected": -216.10397338867188,
      "loss": 0.2744,
      "rewards/accuracies": 0.8812500238418579,
      "rewards/chosen": 2.241814613342285,
      "rewards/margins": 3.3029322624206543,
      "rewards/rejected": -1.0611175298690796,
      "step": 100
    },
    {
      "epoch": 1.4925373134328357,
      "eval_logits/chosen": -2.2504680156707764,
      "eval_logits/rejected": -2.214540958404541,
      "eval_logps/chosen": -258.7962341308594,
      "eval_logps/rejected": -230.7489471435547,
      "eval_loss": 0.582641065120697,
      "eval_rewards/accuracies": 0.7541666626930237,
      "eval_rewards/chosen": 1.7732579708099365,
      "eval_rewards/margins": 2.074618339538574,
      "eval_rewards/rejected": -0.30136018991470337,
      "eval_runtime": 128.1598,
      "eval_samples_per_second": 14.825,
      "eval_steps_per_second": 0.234,
      "step": 100
    },
    {
      "epoch": 1.5671641791044775,
      "grad_norm": 27.34091713721886,
      "learning_rate": 5.041119859162068e-07,
      "logits/chosen": -2.250704288482666,
      "logits/rejected": -2.2283782958984375,
      "logps/chosen": -231.23770141601562,
      "logps/rejected": -248.94601440429688,
      "loss": 0.3566,
      "rewards/accuracies": 0.875,
      "rewards/chosen": 1.9804214239120483,
      "rewards/margins": 3.1362643241882324,
      "rewards/rejected": -1.155842661857605,
      "step": 105
    },
    {
      "epoch": 1.6417910447761193,
      "grad_norm": 20.87748356452889,
      "learning_rate": 4.630254916940423e-07,
      "logits/chosen": -2.261709451675415,
      "logits/rejected": -2.2661125659942627,
      "logps/chosen": -249.21923828125,
      "logps/rejected": -220.63522338867188,
      "loss": 0.264,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": 2.4066946506500244,
      "rewards/margins": 3.436826229095459,
      "rewards/rejected": -1.0301319360733032,
      "step": 110
    },
    {
      "epoch": 1.716417910447761,
      "grad_norm": 22.912621135074613,
      "learning_rate": 4.2218893451814e-07,
      "logits/chosen": -2.292443037033081,
      "logits/rejected": -2.259165048599243,
      "logps/chosen": -233.6214141845703,
      "logps/rejected": -241.71932983398438,
      "loss": 0.2799,
      "rewards/accuracies": 0.9125000238418579,
      "rewards/chosen": 2.4479854106903076,
      "rewards/margins": 3.591277599334717,
      "rewards/rejected": -1.1432924270629883,
      "step": 115
    },
    {
      "epoch": 1.7910447761194028,
      "grad_norm": 24.35363066430913,
      "learning_rate": 3.8187835777481375e-07,
      "logits/chosen": -2.2609424591064453,
      "logits/rejected": -2.2434916496276855,
      "logps/chosen": -248.3112335205078,
      "logps/rejected": -246.8601531982422,
      "loss": 0.2984,
      "rewards/accuracies": 0.887499988079071,
      "rewards/chosen": 2.424971342086792,
      "rewards/margins": 3.6415963172912598,
      "rewards/rejected": -1.2166249752044678,
      "step": 120
    },
    {
      "epoch": 1.8656716417910446,
      "grad_norm": 25.352004598410385,
      "learning_rate": 3.423662493738687e-07,
      "logits/chosen": -2.2195181846618652,
      "logits/rejected": -2.2155442237854004,
      "logps/chosen": -256.672119140625,
      "logps/rejected": -228.56564331054688,
      "loss": 0.3224,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": 2.248502016067505,
      "rewards/margins": 3.5049526691436768,
      "rewards/rejected": -1.2564506530761719,
      "step": 125
    },
    {
      "epoch": 1.9402985074626866,
      "grad_norm": 21.645145170059134,
      "learning_rate": 3.039196998086687e-07,
      "logits/chosen": -2.1958467960357666,
      "logits/rejected": -2.1604387760162354,
      "logps/chosen": -229.57211303710938,
      "logps/rejected": -212.197265625,
      "loss": 0.2811,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 1.9145238399505615,
      "rewards/margins": 3.249605655670166,
      "rewards/rejected": -1.3350815773010254,
      "step": 130
    },
    {
      "epoch": 2.014925373134328,
      "grad_norm": 18.679729678206872,
      "learning_rate": 2.667985967011878e-07,
      "logits/chosen": -2.2026612758636475,
      "logits/rejected": -2.171146869659424,
      "logps/chosen": -232.0654296875,
      "logps/rejected": -220.94296264648438,
      "loss": 0.2681,
      "rewards/accuracies": 0.862500011920929,
      "rewards/chosen": 2.0347559452056885,
      "rewards/margins": 3.320117950439453,
      "rewards/rejected": -1.2853620052337646,
      "step": 135
    },
    {
      "epoch": 2.08955223880597,
      "grad_norm": 15.4753973482206,
      "learning_rate": 2.3125386803640183e-07,
      "logits/chosen": -2.2023463249206543,
      "logits/rejected": -2.1727426052093506,
      "logps/chosen": -234.02597045898438,
      "logps/rejected": -226.4180908203125,
      "loss": 0.1641,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 2.132944107055664,
      "rewards/margins": 3.7610538005828857,
      "rewards/rejected": -1.6281101703643799,
      "step": 140
    },
    {
      "epoch": 2.1641791044776117,
      "grad_norm": 10.981471422963567,
      "learning_rate": 1.9752578596124952e-07,
      "logits/chosen": -2.181061029434204,
      "logits/rejected": -2.157305955886841,
      "logps/chosen": -246.1123809814453,
      "logps/rejected": -220.2969207763672,
      "loss": 0.1649,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 2.136397123336792,
      "rewards/margins": 3.407181978225708,
      "rewards/rejected": -1.2707847356796265,
      "step": 145
    },
    {
      "epoch": 2.2388059701492535,
      "grad_norm": 15.439766803276964,
      "learning_rate": 1.6584234261399532e-07,
      "logits/chosen": -2.183276414871216,
      "logits/rejected": -2.1360280513763428,
      "logps/chosen": -231.65115356445312,
      "logps/rejected": -234.5337371826172,
      "loss": 0.1721,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": 2.034803867340088,
      "rewards/margins": 3.715961456298828,
      "rewards/rejected": -1.6811578273773193,
      "step": 150
    },
    {
      "epoch": 2.2388059701492535,
      "eval_logits/chosen": -2.1955771446228027,
      "eval_logits/rejected": -2.154592275619507,
      "eval_logps/chosen": -262.1832275390625,
      "eval_logps/rejected": -235.352294921875,
      "eval_loss": 0.5628013014793396,
      "eval_rewards/accuracies": 0.7833333611488342,
      "eval_rewards/chosen": 1.434559941291809,
      "eval_rewards/margins": 2.196255922317505,
      "eval_rewards/rejected": -0.761695921421051,
      "eval_runtime": 128.0621,
      "eval_samples_per_second": 14.837,
      "eval_steps_per_second": 0.234,
      "step": 150
    },
    {
      "epoch": 2.3134328358208958,
      "grad_norm": 16.41313227347857,
      "learning_rate": 1.3641770896292082e-07,
      "logits/chosen": -2.1751556396484375,
      "logits/rejected": -2.151604652404785,
      "logps/chosen": -232.66000366210938,
      "logps/rejected": -227.700927734375,
      "loss": 0.1745,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": 2.053588390350342,
      "rewards/margins": 3.8452460765838623,
      "rewards/rejected": -1.7916572093963623,
      "step": 155
    },
    {
      "epoch": 2.388059701492537,
      "grad_norm": 16.889434715439425,
      "learning_rate": 1.0945078707215221e-07,
      "logits/chosen": -2.1910488605499268,
      "logits/rejected": -2.1933162212371826,
      "logps/chosen": -238.77395629882812,
      "logps/rejected": -239.79336547851562,
      "loss": 0.1438,
      "rewards/accuracies": 0.9624999761581421,
      "rewards/chosen": 2.2503340244293213,
      "rewards/margins": 3.826685667037964,
      "rewards/rejected": -1.5763521194458008,
      "step": 160
    },
    {
      "epoch": 2.4626865671641793,
      "grad_norm": 19.03134209250442,
      "learning_rate": 8.512386558088919e-08,
      "logits/chosen": -2.2137415409088135,
      "logits/rejected": -2.1879067420959473,
      "logps/chosen": -238.5040740966797,
      "logps/rejected": -236.75723266601562,
      "loss": 0.162,
      "rewards/accuracies": 0.8687499761581421,
      "rewards/chosen": 2.0795581340789795,
      "rewards/margins": 3.6188082695007324,
      "rewards/rejected": -1.5392506122589111,
      "step": 165
    },
    {
      "epoch": 2.5373134328358207,
      "grad_norm": 22.44154903636031,
      "learning_rate": 6.360138748461013e-08,
      "logits/chosen": -2.2208468914031982,
      "logits/rejected": -2.1851959228515625,
      "logps/chosen": -212.2825164794922,
      "logps/rejected": -229.54971313476562,
      "loss": 0.163,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": 1.9300777912139893,
      "rewards/margins": 3.8205409049987793,
      "rewards/rejected": -1.8904632329940796,
      "step": 170
    },
    {
      "epoch": 2.611940298507463,
      "grad_norm": 15.897018388360651,
      "learning_rate": 4.5028838547699346e-08,
      "logits/chosen": -2.218470573425293,
      "logits/rejected": -2.199413299560547,
      "logps/chosen": -245.3433074951172,
      "logps/rejected": -241.43179321289062,
      "loss": 0.1588,
      "rewards/accuracies": 0.9312499761581421,
      "rewards/chosen": 2.3466992378234863,
      "rewards/margins": 4.257248878479004,
      "rewards/rejected": -1.9105498790740967,
      "step": 175
    },
    {
      "epoch": 2.6865671641791042,
      "grad_norm": 19.837936021486545,
      "learning_rate": 2.9531763861505964e-08,
      "logits/chosen": -2.1986160278320312,
      "logits/rejected": -2.1841235160827637,
      "logps/chosen": -231.4283905029297,
      "logps/rejected": -225.1343231201172,
      "loss": 0.147,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 2.088632822036743,
      "rewards/margins": 3.8668549060821533,
      "rewards/rejected": -1.7782217264175415,
      "step": 180
    },
    {
      "epoch": 2.7611940298507465,
      "grad_norm": 18.919043908593903,
      "learning_rate": 1.7214919195619125e-08,
      "logits/chosen": -2.1973977088928223,
      "logits/rejected": -2.180459499359131,
      "logps/chosen": -241.5362091064453,
      "logps/rejected": -247.6551513671875,
      "loss": 0.154,
      "rewards/accuracies": 0.9437500238418579,
      "rewards/chosen": 2.1070444583892822,
      "rewards/margins": 3.8555946350097656,
      "rewards/rejected": -1.7485500574111938,
      "step": 185
    },
    {
      "epoch": 2.835820895522388,
      "grad_norm": 15.256609349963595,
      "learning_rate": 8.161562878982398e-09,
      "logits/chosen": -2.187682628631592,
      "logits/rejected": -2.1831722259521484,
      "logps/chosen": -241.7909393310547,
      "logps/rejected": -239.01785278320312,
      "loss": 0.1512,
      "rewards/accuracies": 0.925000011920929,
      "rewards/chosen": 2.381829261779785,
      "rewards/margins": 4.050623416900635,
      "rewards/rejected": -1.66879403591156,
      "step": 190
    },
    {
      "epoch": 2.91044776119403,
      "grad_norm": 14.520596186258233,
      "learning_rate": 2.432892997526026e-09,
      "logits/chosen": -2.210559368133545,
      "logits/rejected": -2.183519124984741,
      "logps/chosen": -254.4117889404297,
      "logps/rejected": -243.2581787109375,
      "loss": 0.1529,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": 2.8705074787139893,
      "rewards/margins": 4.513058662414551,
      "rewards/rejected": -1.6425514221191406,
      "step": 195
    },
    {
      "epoch": 2.9850746268656714,
      "grad_norm": 23.157831299488215,
      "learning_rate": 6.763371270035457e-11,
      "logits/chosen": -2.211535692214966,
      "logits/rejected": -2.1897168159484863,
      "logps/chosen": -225.7974395751953,
      "logps/rejected": -239.18331909179688,
      "loss": 0.1926,
      "rewards/accuracies": 0.918749988079071,
      "rewards/chosen": 2.324434280395508,
      "rewards/margins": 3.6570217609405518,
      "rewards/rejected": -1.332587718963623,
      "step": 200
    },
    {
      "epoch": 2.9850746268656714,
      "eval_logits/chosen": -2.2166454792022705,
      "eval_logits/rejected": -2.176008701324463,
      "eval_logps/chosen": -260.0589904785156,
      "eval_logps/rejected": -234.689208984375,
      "eval_loss": 0.5716387629508972,
      "eval_rewards/accuracies": 0.7875000238418579,
      "eval_rewards/chosen": 1.6469833850860596,
      "eval_rewards/margins": 2.342369556427002,
      "eval_rewards/rejected": -0.6953860521316528,
      "eval_runtime": 128.0997,
      "eval_samples_per_second": 14.832,
      "eval_steps_per_second": 0.234,
      "step": 200
    }
  ],
  "logging_steps": 5,
  "max_steps": 201,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 2369906314051584.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}