{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.992,
  "eval_steps": 100,
  "global_step": 124,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 3.846153846153847e-07,
      "logits/chosen": 0.7132297158241272,
      "logits/rejected": 0.7902912497520447,
      "logps/chosen": -182.8937530517578,
      "logps/rejected": -166.20700073242188,
      "loss": 0.0102,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.846153846153847e-06,
      "logits/chosen": 0.87026047706604,
      "logits/rejected": 0.8367093801498413,
      "logps/chosen": -155.32986450195312,
      "logps/rejected": -139.0651397705078,
      "loss": 0.0102,
      "rewards/accuracies": 0.3611111044883728,
      "rewards/chosen": 0.0006189693231135607,
      "rewards/margins": 9.571156988386065e-05,
      "rewards/rejected": 0.0005232577677816153,
      "step": 10
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.951096619903317e-06,
      "logits/chosen": 0.8673034906387329,
      "logits/rejected": 0.8491884469985962,
      "logps/chosen": -148.64974975585938,
      "logps/rejected": -141.98843383789062,
      "loss": 0.0103,
      "rewards/accuracies": 0.33125001192092896,
      "rewards/chosen": 6.371500785462558e-05,
      "rewards/margins": -0.00015192512364592403,
      "rewards/rejected": 0.00021564005874097347,
      "step": 20
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.716164218065246e-06,
      "logits/chosen": 0.8203651309013367,
      "logits/rejected": 0.8144901990890503,
      "logps/chosen": -143.59263610839844,
      "logps/rejected": -129.68614196777344,
      "loss": 0.0102,
      "rewards/accuracies": 0.4124999940395355,
      "rewards/chosen": 0.0005150804063305259,
      "rewards/margins": 0.00021757767535746098,
      "rewards/rejected": 0.00029750276007689536,
      "step": 30
    },
    {
      "epoch": 0.64,
      "learning_rate": 4.3048902348863116e-06,
      "logits/chosen": 0.8285233378410339,
      "logits/rejected": 0.8196646571159363,
      "logps/chosen": -175.5411376953125,
      "logps/rejected": -162.92697143554688,
      "loss": 0.0104,
      "rewards/accuracies": 0.3187499940395355,
      "rewards/chosen": -3.423704811211792e-06,
      "rewards/margins": -0.0005603213212452829,
      "rewards/rejected": 0.0005568976048380136,
      "step": 40
    },
    {
      "epoch": 0.8,
      "learning_rate": 3.7500000000000005e-06,
      "logits/chosen": 0.8533094525337219,
      "logits/rejected": 0.8660014271736145,
      "logps/chosen": -154.22348022460938,
      "logps/rejected": -141.3745880126953,
      "loss": 0.0102,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": -0.0006007859483361244,
      "rewards/margins": 9.4042276032269e-05,
      "rewards/rejected": -0.0006948282243683934,
      "step": 50
    },
    {
      "epoch": 0.96,
      "learning_rate": 3.0956464785579125e-06,
      "logits/chosen": 0.840475857257843,
      "logits/rejected": 0.835978627204895,
      "logps/chosen": -147.8980255126953,
      "logps/rejected": -135.54531860351562,
      "loss": 0.0102,
      "rewards/accuracies": 0.3499999940395355,
      "rewards/chosen": -0.0003634719760157168,
      "rewards/margins": -3.2313815609086305e-05,
      "rewards/rejected": -0.00033115819678641856,
      "step": 60
    },
    {
      "epoch": 1.12,
      "learning_rate": 2.39389699200963e-06,
      "logits/chosen": 0.9384347796440125,
      "logits/rejected": 0.9138472676277161,
      "logps/chosen": -141.71607971191406,
      "logps/rejected": -131.02381896972656,
      "loss": 0.0102,
      "rewards/accuracies": 0.4000000059604645,
      "rewards/chosen": -0.0002880954125430435,
      "rewards/margins": 0.0002658166631590575,
      "rewards/rejected": -0.0005539121339097619,
      "step": 70
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.700590188571887e-06,
      "logits/chosen": 0.8153573274612427,
      "logits/rejected": 0.8099604845046997,
      "logps/chosen": -133.85093688964844,
      "logps/rejected": -118.66615295410156,
      "loss": 0.0102,
      "rewards/accuracies": 0.41874998807907104,
      "rewards/chosen": -0.00017777756147552282,
      "rewards/margins": 0.0004668553010560572,
      "rewards/rejected": -0.000644632731564343,
      "step": 80
    },
    {
      "epoch": 1.44,
      "learning_rate": 1.0708929268538034e-06,
      "logits/chosen": 0.8284982442855835,
      "logits/rejected": 0.8237984776496887,
      "logps/chosen": -158.91497802734375,
      "logps/rejected": -146.9730682373047,
      "loss": 0.0102,
      "rewards/accuracies": 0.4437499940395355,
      "rewards/chosen": -0.000682778307236731,
      "rewards/margins": 0.0004111085145268589,
      "rewards/rejected": -0.0010938867926597595,
      "step": 90
    },
    {
      "epoch": 1.6,
      "learning_rate": 5.549106142039018e-07,
      "logits/chosen": 0.8617427945137024,
      "logits/rejected": 0.8781857490539551,
      "logps/chosen": -152.7745819091797,
      "logps/rejected": -141.37179565429688,
      "loss": 0.0103,
      "rewards/accuracies": 0.34375,
      "rewards/chosen": -0.001233941875398159,
      "rewards/margins": -0.00010122207459062338,
      "rewards/rejected": -0.0011327198008075356,
      "step": 100
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.937002879188285e-07,
      "logits/chosen": 0.8532508611679077,
      "logits/rejected": 0.8138865232467651,
      "logps/chosen": -175.2100830078125,
      "logps/rejected": -158.42974853515625,
      "loss": 0.0102,
      "rewards/accuracies": 0.48124998807907104,
      "rewards/chosen": -0.0008482037228532135,
      "rewards/margins": 0.00030287052504718304,
      "rewards/rejected": -0.0011510740732774138,
      "step": 110
    },
    {
      "epoch": 1.93,
      "learning_rate": 1.6003680950742728e-08,
      "logits/chosen": 0.8294161558151245,
      "logits/rejected": 0.826370120048523,
      "logps/chosen": -158.0736083984375,
      "logps/rejected": -146.1743927001953,
      "loss": 0.01,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.0006492567481473088,
      "rewards/margins": 0.0010562509996816516,
      "rewards/rejected": -0.0017055077478289604,
      "step": 120
    },
    {
      "epoch": 1.99,
      "step": 124,
      "total_flos": 0.0,
      "train_loss": 0.001956978782771095,
      "train_runtime": 290.9594,
      "train_samples_per_second": 6.874,
      "train_steps_per_second": 0.426
    }
  ],
  "logging_steps": 10,
  "max_steps": 124,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}