{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 500,
  "global_step": 156,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 16.144156476964564,
      "learning_rate": 3.125e-08,
      "logits/chosen": -1.517999529838562,
      "logits/rejected": -1.427964687347412,
      "logps/chosen": -138.13075256347656,
      "logps/rejected": -139.19334411621094,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "grad_norm": 16.23747042290227,
      "learning_rate": 3.1249999999999997e-07,
      "logits/chosen": -1.8612151145935059,
      "logits/rejected": -1.837838053703308,
      "logps/chosen": -165.77423095703125,
      "logps/rejected": -167.00115966796875,
      "loss": 0.6932,
      "rewards/accuracies": 0.4097222089767456,
      "rewards/chosen": 0.00027224430232308805,
      "rewards/margins": -0.0006895032711327076,
      "rewards/rejected": 0.0009617475443519652,
      "step": 10
    },
    {
      "epoch": 0.13,
      "grad_norm": 16.626258757403523,
      "learning_rate": 4.989935734988097e-07,
      "logits/chosen": -1.765881896018982,
      "logits/rejected": -1.758798360824585,
      "logps/chosen": -146.74258422851562,
      "logps/rejected": -150.7355499267578,
      "loss": 0.6929,
      "rewards/accuracies": 0.503125011920929,
      "rewards/chosen": -0.013516152277588844,
      "rewards/margins": 0.0023722327314317226,
      "rewards/rejected": -0.015888383612036705,
      "step": 20
    },
    {
      "epoch": 0.19,
      "grad_norm": 15.590271696523773,
      "learning_rate": 4.877641290737883e-07,
      "logits/chosen": -1.6886975765228271,
      "logits/rejected": -1.7255547046661377,
      "logps/chosen": -171.6064453125,
      "logps/rejected": -175.8733367919922,
      "loss": 0.6936,
      "rewards/accuracies": 0.4781250059604645,
      "rewards/chosen": -0.10228633880615234,
      "rewards/margins": -0.003102194517850876,
      "rewards/rejected": -0.09918414056301117,
      "step": 30
    },
    {
      "epoch": 0.26,
      "grad_norm": 15.611757303970759,
      "learning_rate": 4.646121984004665e-07,
      "logits/chosen": -1.6890376806259155,
      "logits/rejected": -1.6796882152557373,
      "logps/chosen": -175.96786499023438,
      "logps/rejected": -174.4598846435547,
      "loss": 0.6962,
      "rewards/accuracies": 0.44062501192092896,
      "rewards/chosen": -0.1981964409351349,
      "rewards/margins": -0.007378303911536932,
      "rewards/rejected": -0.19081811606884003,
      "step": 40
    },
    {
      "epoch": 0.32,
      "grad_norm": 15.152656997490112,
      "learning_rate": 4.3069871595684787e-07,
      "logits/chosen": -1.8153579235076904,
      "logits/rejected": -1.864524483680725,
      "logps/chosen": -173.35693359375,
      "logps/rejected": -175.8848114013672,
      "loss": 0.6916,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": -0.09905640780925751,
      "rewards/margins": 0.00832386501133442,
      "rewards/rejected": -0.10738028585910797,
      "step": 50
    },
    {
      "epoch": 0.38,
      "grad_norm": 17.022534458266477,
      "learning_rate": 3.877242453630256e-07,
      "logits/chosen": -1.8550310134887695,
      "logits/rejected": -1.8582820892333984,
      "logps/chosen": -192.25155639648438,
      "logps/rejected": -197.19004821777344,
      "loss": 0.6912,
      "rewards/accuracies": 0.518750011920929,
      "rewards/chosen": -0.2872371971607208,
      "rewards/margins": 0.02158377319574356,
      "rewards/rejected": -0.3088209331035614,
      "step": 60
    },
    {
      "epoch": 0.45,
      "grad_norm": 16.571326277539317,
      "learning_rate": 3.378437060203357e-07,
      "logits/chosen": -1.8471654653549194,
      "logits/rejected": -1.8384788036346436,
      "logps/chosen": -192.4368896484375,
      "logps/rejected": -195.90115356445312,
      "loss": 0.6888,
      "rewards/accuracies": 0.515625,
      "rewards/chosen": -0.3420848548412323,
      "rewards/margins": 0.01330840028822422,
      "rewards/rejected": -0.35539326071739197,
      "step": 70
    },
    {
      "epoch": 0.51,
      "grad_norm": 13.438756015994942,
      "learning_rate": 2.8355831645441387e-07,
      "logits/chosen": -1.8971712589263916,
      "logits/rejected": -1.9240859746932983,
      "logps/chosen": -172.50802612304688,
      "logps/rejected": -181.87484741210938,
      "loss": 0.6872,
      "rewards/accuracies": 0.5218750238418579,
      "rewards/chosen": -0.19086754322052002,
      "rewards/margins": 0.0247647762298584,
      "rewards/rejected": -0.21563228964805603,
      "step": 80
    },
    {
      "epoch": 0.58,
      "grad_norm": 15.835235204252495,
      "learning_rate": 2.2759017277414164e-07,
      "logits/chosen": -1.803915023803711,
      "logits/rejected": -1.7513706684112549,
      "logps/chosen": -196.07321166992188,
      "logps/rejected": -192.37071228027344,
      "loss": 0.683,
      "rewards/accuracies": 0.5531250238418579,
      "rewards/chosen": -0.3334119915962219,
      "rewards/margins": 0.01788436248898506,
      "rewards/rejected": -0.3512963652610779,
      "step": 90
    },
    {
      "epoch": 0.64,
      "grad_norm": 20.64165185557883,
      "learning_rate": 1.7274575140626315e-07,
      "logits/chosen": -1.8089154958724976,
      "logits/rejected": -1.7900733947753906,
      "logps/chosen": -193.5570831298828,
      "logps/rejected": -195.0845489501953,
      "loss": 0.6909,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.3730206787586212,
      "rewards/margins": 0.006149230990558863,
      "rewards/rejected": -0.37916988134384155,
      "step": 100
    },
    {
      "epoch": 0.7,
      "grad_norm": 14.009691318366311,
      "learning_rate": 1.2177518064852348e-07,
      "logits/chosen": -1.735630750656128,
      "logits/rejected": -1.7698205709457397,
      "logps/chosen": -179.77761840820312,
      "logps/rejected": -190.6461944580078,
      "loss": 0.6853,
      "rewards/accuracies": 0.596875011920929,
      "rewards/chosen": -0.3420836329460144,
      "rewards/margins": 0.045021723955869675,
      "rewards/rejected": -0.38710540533065796,
      "step": 110
    },
    {
      "epoch": 0.77,
      "grad_norm": 15.208350623463305,
      "learning_rate": 7.723433775328384e-08,
      "logits/chosen": -1.8086316585540771,
      "logits/rejected": -1.8518075942993164,
      "logps/chosen": -198.0193634033203,
      "logps/rejected": -197.88845825195312,
      "loss": 0.6845,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.36722415685653687,
      "rewards/margins": 0.019394848495721817,
      "rewards/rejected": -0.386618971824646,
      "step": 120
    },
    {
      "epoch": 0.83,
      "grad_norm": 20.27285082098782,
      "learning_rate": 4.1356686569674335e-08,
      "logits/chosen": -1.6780946254730225,
      "logits/rejected": -1.6494309902191162,
      "logps/chosen": -181.09170532226562,
      "logps/rejected": -182.8488311767578,
      "loss": 0.6846,
      "rewards/accuracies": 0.5687500238418579,
      "rewards/chosen": -0.29602062702178955,
      "rewards/margins": 0.0265937689691782,
      "rewards/rejected": -0.3226144015789032,
      "step": 130
    },
    {
      "epoch": 0.9,
      "grad_norm": 16.756793414757478,
      "learning_rate": 1.5941282340065697e-08,
      "logits/chosen": -1.8084728717803955,
      "logits/rejected": -1.767961859703064,
      "logps/chosen": -189.6092071533203,
      "logps/rejected": -193.9722900390625,
      "loss": 0.6818,
      "rewards/accuracies": 0.543749988079071,
      "rewards/chosen": -0.3253946006298065,
      "rewards/margins": 0.02885589934885502,
      "rewards/rejected": -0.3542505204677582,
      "step": 140
    },
    {
      "epoch": 0.96,
      "grad_norm": 15.87524427688872,
      "learning_rate": 2.2625595580163247e-09,
      "logits/chosen": -1.70889413356781,
      "logits/rejected": -1.748719573020935,
      "logps/chosen": -189.88803100585938,
      "logps/rejected": -193.5582733154297,
      "loss": 0.6852,
      "rewards/accuracies": 0.565625011920929,
      "rewards/chosen": -0.38032156229019165,
      "rewards/margins": 0.030102457851171494,
      "rewards/rejected": -0.41042399406433105,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 156,
      "total_flos": 0.0,
      "train_loss": 0.688686496936358,
      "train_runtime": 18523.9468,
      "train_samples_per_second": 1.08,
      "train_steps_per_second": 0.008
    }
  ],
  "logging_steps": 10,
  "max_steps": 156,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}