{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9990817263544536,
  "eval_steps": 100,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 2.3809523809523811e-07,
      "logits/chosen": -2.530543327331543,
      "logits/rejected": -2.3806393146514893,
      "logps/chosen": -277.43206787109375,
      "logps/rejected": -259.5929870605469,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.380952380952381e-06,
      "logits/chosen": -2.3367958068847656,
      "logits/rejected": -2.3201417922973633,
      "logps/chosen": -237.50930786132812,
      "logps/rejected": -259.6923828125,
      "loss": 0.6911,
      "rewards/accuracies": 0.5138888955116272,
      "rewards/chosen": -0.0002860191452782601,
      "rewards/margins": 0.004583699628710747,
      "rewards/rejected": -0.004869719035923481,
      "step": 10
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.761904761904762e-06,
      "logits/chosen": -2.2300033569335938,
      "logits/rejected": -2.062291383743286,
      "logps/chosen": -322.0083312988281,
      "logps/rejected": -298.00164794921875,
      "loss": 0.6623,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.05355127900838852,
      "rewards/margins": 0.07823377847671509,
      "rewards/rejected": -0.1317850649356842,
      "step": 20
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.970219740227693e-06,
      "logits/chosen": -1.7139427661895752,
      "logits/rejected": -1.5317604541778564,
      "logps/chosen": -302.0789489746094,
      "logps/rejected": -304.89666748046875,
      "loss": 0.6183,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.13164304196834564,
      "rewards/margins": 0.1831807643175125,
      "rewards/rejected": -0.31482380628585815,
      "step": 30
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.868186180746792e-06,
      "logits/chosen": -1.342703104019165,
      "logits/rejected": -1.0153623819351196,
      "logps/chosen": -317.6343688964844,
      "logps/rejected": -340.515625,
      "loss": 0.6019,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.03232262283563614,
      "rewards/margins": 0.30836719274520874,
      "rewards/rejected": -0.34068983793258667,
      "step": 40
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.696530612642871e-06,
      "logits/chosen": -0.4007461667060852,
      "logits/rejected": -0.062369205057621,
      "logps/chosen": -344.5245056152344,
      "logps/rejected": -381.02655029296875,
      "loss": 0.5788,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.4034351408481598,
      "rewards/margins": 0.35783329606056213,
      "rewards/rejected": -0.7612683773040771,
      "step": 50
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.460299516441777e-06,
      "logits/chosen": -0.4702862799167633,
      "logits/rejected": -0.10355714708566666,
      "logps/chosen": -309.2970886230469,
      "logps/rejected": -356.2215881347656,
      "loss": 0.5714,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": -0.17937391996383667,
      "rewards/margins": 0.42016029357910156,
      "rewards/rejected": -0.5995342135429382,
      "step": 60
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.1664378205239085e-06,
      "logits/chosen": -0.14899344742298126,
      "logits/rejected": 0.4177340567111969,
      "logps/chosen": -340.29766845703125,
      "logps/rejected": -379.6929626464844,
      "loss": 0.5634,
      "rewards/accuracies": 0.684374988079071,
      "rewards/chosen": -0.4960289001464844,
      "rewards/margins": 0.4510241448879242,
      "rewards/rejected": -0.9470531344413757,
      "step": 70
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.8235847280454626e-06,
      "logits/chosen": -0.14341413974761963,
      "logits/rejected": 0.4753722548484802,
      "logps/chosen": -334.1756286621094,
      "logps/rejected": -374.01556396484375,
      "loss": 0.527,
      "rewards/accuracies": 0.715624988079071,
      "rewards/chosen": -0.3731788992881775,
      "rewards/margins": 0.5048251152038574,
      "rewards/rejected": -0.8780039548873901,
      "step": 80
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.441819734087963e-06,
      "logits/chosen": -0.20099422335624695,
      "logits/rejected": 0.4678555130958557,
      "logps/chosen": -348.76397705078125,
      "logps/rejected": -384.9765319824219,
      "loss": 0.5453,
      "rewards/accuracies": 0.7406250238418579,
      "rewards/chosen": -0.3222987651824951,
      "rewards/margins": 0.5874095559120178,
      "rewards/rejected": -0.9097083210945129,
      "step": 90
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.0323662998460396e-06,
      "logits/chosen": -0.17653772234916687,
      "logits/rejected": 0.5543693900108337,
      "logps/chosen": -353.01361083984375,
      "logps/rejected": -381.8619689941406,
      "loss": 0.5359,
      "rewards/accuracies": 0.703125,
      "rewards/chosen": -0.41476544737815857,
      "rewards/margins": 0.5261854529380798,
      "rewards/rejected": -0.940950870513916,
      "step": 100
    },
    {
      "epoch": 0.49,
      "eval_logits/chosen": -0.04770730063319206,
      "eval_logits/rejected": 0.5653438568115234,
      "eval_logps/chosen": -348.3134765625,
      "eval_logps/rejected": -383.2900695800781,
      "eval_loss": 0.5605703592300415,
      "eval_rewards/accuracies": 0.699999988079071,
      "eval_rewards/chosen": -0.4171307384967804,
      "eval_rewards/margins": 0.5390676259994507,
      "eval_rewards/rejected": -0.9561984539031982,
      "eval_runtime": 383.9702,
      "eval_samples_per_second": 5.209,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.6072618954988867e-06,
      "logits/chosen": -0.024105533957481384,
      "logits/rejected": 0.6808269619941711,
      "logps/chosen": -357.918701171875,
      "logps/rejected": -408.1763610839844,
      "loss": 0.5632,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.5132647752761841,
      "rewards/margins": 0.5779321193695068,
      "rewards/rejected": -1.0911967754364014,
      "step": 110
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1790041121336223e-06,
      "logits/chosen": -0.00899410154670477,
      "logits/rejected": 0.5235549211502075,
      "logps/chosen": -369.63299560546875,
      "logps/rejected": -408.86859130859375,
      "loss": 0.5253,
      "rewards/accuracies": 0.7281249761581421,
      "rewards/chosen": -0.640425980091095,
      "rewards/margins": 0.5291002988815308,
      "rewards/rejected": -1.1695263385772705,
      "step": 120
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.760183246631777e-06,
      "logits/chosen": 0.018746722489595413,
      "logits/rejected": 0.6442402005195618,
      "logps/chosen": -351.2228088378906,
      "logps/rejected": -409.70623779296875,
      "loss": 0.5308,
      "rewards/accuracies": 0.7281249761581421,
      "rewards/chosen": -0.4084150195121765,
      "rewards/margins": 0.7148237824440002,
      "rewards/rejected": -1.1232386827468872,
      "step": 130
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3631121611097364e-06,
      "logits/chosen": 0.052795905619859695,
      "logits/rejected": 0.5363299250602722,
      "logps/chosen": -350.0050354003906,
      "logps/rejected": -389.28875732421875,
      "loss": 0.5368,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.411121129989624,
      "rewards/margins": 0.5764644742012024,
      "rewards/rejected": -0.9875855445861816,
      "step": 140
    },
    {
      "epoch": 0.73,
      "learning_rate": 9.994642986290797e-07,
      "logits/chosen": -0.08179731667041779,
      "logits/rejected": 0.6078404784202576,
      "logps/chosen": -340.8910217285156,
      "logps/rejected": -395.11309814453125,
      "loss": 0.5396,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.4558540880680084,
      "rewards/margins": 0.681885302066803,
      "rewards/rejected": -1.1377394199371338,
      "step": 150
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.799304971075383e-07,
      "logits/chosen": -0.10812912881374359,
      "logits/rejected": 0.7408884763717651,
      "logps/chosen": -367.9963684082031,
      "logps/rejected": -409.3976135253906,
      "loss": 0.5303,
      "rewards/accuracies": 0.7406250238418579,
      "rewards/chosen": -0.5040684938430786,
      "rewards/margins": 0.6103622317314148,
      "rewards/rejected": -1.1144306659698486,
      "step": 160
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1390469071538183e-07,
      "logits/chosen": 0.11164654791355133,
      "logits/rejected": 0.6015251278877258,
      "logps/chosen": -354.7975158691406,
      "logps/rejected": -422.2373046875,
      "loss": 0.5308,
      "rewards/accuracies": 0.7562500238418579,
      "rewards/chosen": -0.47427788376808167,
      "rewards/margins": 0.8326796293258667,
      "rewards/rejected": -1.306957483291626,
      "step": 170
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.092077387824884e-07,
      "logits/chosen": -0.10529808700084686,
      "logits/rejected": 0.4767494201660156,
      "logps/chosen": -361.30084228515625,
      "logps/rejected": -419.2337951660156,
      "loss": 0.5424,
      "rewards/accuracies": 0.7093750238418579,
      "rewards/chosen": -0.48817354440689087,
      "rewards/margins": 0.6789232492446899,
      "rewards/rejected": -1.167096734046936,
      "step": 180
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.185750133542168e-08,
      "logits/chosen": 0.029346108436584473,
      "logits/rejected": 0.39755165576934814,
      "logps/chosen": -337.2845764160156,
      "logps/rejected": -398.9493713378906,
      "loss": 0.54,
      "rewards/accuracies": 0.703125,
      "rewards/chosen": -0.5243724584579468,
      "rewards/margins": 0.6791526675224304,
      "rewards/rejected": -1.203525185585022,
      "step": 190
    },
    {
      "epoch": 0.98,
      "learning_rate": 5.891920784984184e-09,
      "logits/chosen": -0.1364443600177765,
      "logits/rejected": 0.5197805166244507,
      "logps/chosen": -351.525390625,
      "logps/rejected": -410.6580505371094,
      "loss": 0.5121,
      "rewards/accuracies": 0.778124988079071,
      "rewards/chosen": -0.4877961575984955,
      "rewards/margins": 0.7241436839103699,
      "rewards/rejected": -1.211940050125122,
      "step": 200
    },
    {
      "epoch": 0.98,
      "eval_logits/chosen": -0.05626865103840828,
      "eval_logits/rejected": 0.5949223637580872,
      "eval_logps/chosen": -360.6636657714844,
      "eval_logps/rejected": -399.93438720703125,
      "eval_loss": 0.5482509732246399,
      "eval_rewards/accuracies": 0.6959999799728394,
      "eval_rewards/chosen": -0.5406328439712524,
      "eval_rewards/margins": 0.5820090770721436,
      "eval_rewards/rejected": -1.122641921043396,
      "eval_runtime": 384.3787,
      "eval_samples_per_second": 5.203,
      "eval_steps_per_second": 0.65,
      "step": 200
    },
    {
      "epoch": 1.0,
      "step": 204,
      "total_flos": 0.0,
      "train_loss": 0.5619814822486803,
      "train_runtime": 9621.729,
      "train_samples_per_second": 2.716,
      "train_steps_per_second": 0.021
    }
  ],
  "logging_steps": 10,
  "max_steps": 204,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}