{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.998324958123953,
  "eval_steps": 100,
  "global_step": 149,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.3333333333333335e-07,
      "logits/chosen": -2.55656099319458,
      "logits/rejected": -2.55704402923584,
      "logps/chosen": -162.36532592773438,
      "logps/rejected": -172.43312072753906,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.3333333333333333e-06,
      "logits/chosen": -2.666642189025879,
      "logits/rejected": -2.6329586505889893,
      "logps/chosen": -185.80641174316406,
      "logps/rejected": -184.18959045410156,
      "loss": 0.6928,
      "rewards/accuracies": 0.4548611044883728,
      "rewards/chosen": -0.001971459249034524,
      "rewards/margins": 0.0016002749325707555,
      "rewards/rejected": -0.003571733832359314,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.982842942906386e-06,
      "logits/chosen": -2.721466541290283,
      "logits/rejected": -2.67402982711792,
      "logps/chosen": -201.33470153808594,
      "logps/rejected": -194.63198852539062,
      "loss": 0.6949,
      "rewards/accuracies": 0.534375011920929,
      "rewards/chosen": -0.10967914760112762,
      "rewards/margins": 0.0006104880012571812,
      "rewards/rejected": -0.11028961837291718,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.846996204000967e-06,
      "logits/chosen": -2.625549077987671,
      "logits/rejected": -2.607099771499634,
      "logps/chosen": -197.00973510742188,
      "logps/rejected": -195.684326171875,
      "loss": 0.6939,
      "rewards/accuracies": 0.5093749761581421,
      "rewards/chosen": -0.08795476704835892,
      "rewards/margins": -0.002985857194289565,
      "rewards/rejected": -0.08496890217065811,
      "step": 30
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.582735470385229e-06,
      "logits/chosen": -2.633084535598755,
      "logits/rejected": -2.634359836578369,
      "logps/chosen": -185.71023559570312,
      "logps/rejected": -185.03509521484375,
      "loss": 0.6921,
      "rewards/accuracies": 0.5531250238418579,
      "rewards/chosen": -0.04532230272889137,
      "rewards/margins": 0.0011206632480025291,
      "rewards/rejected": -0.046442966908216476,
      "step": 40
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.204519553876095e-06,
      "logits/chosen": -2.6109485626220703,
      "logits/rejected": -2.6062893867492676,
      "logps/chosen": -171.55734252929688,
      "logps/rejected": -173.1212158203125,
      "loss": 0.6912,
      "rewards/accuracies": 0.4625000059604645,
      "rewards/chosen": -0.014199512079358101,
      "rewards/margins": 0.0028473488055169582,
      "rewards/rejected": -0.017046859487891197,
      "step": 50
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.7330422317447686e-06,
      "logits/chosen": -2.6457674503326416,
      "logits/rejected": -2.6292178630828857,
      "logps/chosen": -188.72109985351562,
      "logps/rejected": -184.30081176757812,
      "loss": 0.6905,
      "rewards/accuracies": 0.5375000238418579,
      "rewards/chosen": -0.00934157706797123,
      "rewards/margins": 0.002996337367221713,
      "rewards/rejected": -0.012337915599346161,
      "step": 60
    },
    {
      "epoch": 0.47,
      "learning_rate": 3.1941000034687516e-06,
      "logits/chosen": -2.6249117851257324,
      "logits/rejected": -2.611786365509033,
      "logps/chosen": -170.21836853027344,
      "logps/rejected": -177.2921142578125,
      "loss": 0.6905,
      "rewards/accuracies": 0.534375011920929,
      "rewards/chosen": -0.03585924953222275,
      "rewards/margins": 0.0028343182057142258,
      "rewards/rejected": -0.03869356960058212,
      "step": 70
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.6171806561748503e-06,
      "logits/chosen": -2.5946967601776123,
      "logits/rejected": -2.571646213531494,
      "logps/chosen": -184.07911682128906,
      "logps/rejected": -189.4521942138672,
      "loss": 0.6896,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.019358564168214798,
      "rewards/margins": 0.011903460137546062,
      "rewards/rejected": -0.031262025237083435,
      "step": 80
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.0338498642707977e-06,
      "logits/chosen": -2.629333972930908,
      "logits/rejected": -2.6070916652679443,
      "logps/chosen": -183.12803649902344,
      "logps/rejected": -187.22447204589844,
      "loss": 0.6902,
      "rewards/accuracies": 0.565625011920929,
      "rewards/chosen": -0.0804746001958847,
      "rewards/margins": 0.01135367900133133,
      "rewards/rejected": -0.09182827174663544,
      "step": 90
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4760240991587338e-06,
      "logits/chosen": -2.5863940715789795,
      "logits/rejected": -2.5969200134277344,
      "logps/chosen": -172.47142028808594,
      "logps/rejected": -180.51638793945312,
      "loss": 0.6889,
      "rewards/accuracies": 0.5531250238418579,
      "rewards/chosen": -0.05926589295268059,
      "rewards/margins": 0.011102231219410896,
      "rewards/rejected": -0.07036812603473663,
      "step": 100
    },
    {
      "epoch": 0.67,
      "eval_logits/chosen": -2.5383808612823486,
      "eval_logits/rejected": -2.4414656162261963,
      "eval_logps/chosen": -307.9212646484375,
      "eval_logps/rejected": -301.2935485839844,
      "eval_loss": 0.6852481365203857,
      "eval_rewards/accuracies": 0.5659999847412109,
      "eval_rewards/chosen": -0.0951852798461914,
      "eval_rewards/margins": 0.019401030614972115,
      "eval_rewards/rejected": -0.11458631604909897,
      "eval_runtime": 382.181,
      "eval_samples_per_second": 5.233,
      "eval_steps_per_second": 0.654,
      "step": 100
    },
    {
      "epoch": 0.74,
      "learning_rate": 9.742243453755202e-07,
      "logits/chosen": -2.612968683242798,
      "logits/rejected": -2.6128811836242676,
      "logps/chosen": -176.701416015625,
      "logps/rejected": -186.86849975585938,
      "loss": 0.6882,
      "rewards/accuracies": 0.5406249761581421,
      "rewards/chosen": -0.027405280619859695,
      "rewards/margins": 0.016830626875162125,
      "rewards/rejected": -0.04423590749502182,
      "step": 110
    },
    {
      "epoch": 0.8,
      "learning_rate": 5.559061696656199e-07,
      "logits/chosen": -2.585603713989258,
      "logits/rejected": -2.5767314434051514,
      "logps/chosen": -184.33139038085938,
      "logps/rejected": -189.44131469726562,
      "loss": 0.686,
      "rewards/accuracies": 0.6031249761581421,
      "rewards/chosen": -0.03763353452086449,
      "rewards/margins": 0.025313779711723328,
      "rewards/rejected": -0.06294731795787811,
      "step": 120
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.4395751190352924e-07,
      "logits/chosen": -2.5703272819519043,
      "logits/rejected": -2.5677199363708496,
      "logps/chosen": -184.06341552734375,
      "logps/rejected": -188.58779907226562,
      "loss": 0.6884,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.06450501084327698,
      "rewards/margins": 0.007478479295969009,
      "rewards/rejected": -0.07198350131511688,
      "step": 130
    },
    {
      "epoch": 0.94,
      "learning_rate": 5.544639001763719e-08,
      "logits/chosen": -2.6024391651153564,
      "logits/rejected": -2.5870203971862793,
      "logps/chosen": -199.04031372070312,
      "logps/rejected": -197.60330200195312,
      "loss": 0.686,
      "rewards/accuracies": 0.565625011920929,
      "rewards/chosen": -0.06172681599855423,
      "rewards/margins": 0.017358267679810524,
      "rewards/rejected": -0.0790850818157196,
      "step": 140
    },
    {
      "epoch": 1.0,
      "step": 149,
      "total_flos": 0.0,
      "train_loss": 0.6901595840518107,
      "train_runtime": 6945.5042,
      "train_samples_per_second": 2.75,
      "train_steps_per_second": 0.021
    }
  ],
  "logging_steps": 10,
  "max_steps": 149,
  "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
}