{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9965156794425087,
  "eval_steps": 50,
  "global_step": 143,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06968641114982578,
      "grad_norm": 7.952149336949629,
      "learning_rate": 3.333333333333333e-07,
      "logits/chosen": -2.7471699714660645,
      "logits/rejected": -2.7366185188293457,
      "logps/chosen": -291.8603820800781,
      "logps/rejected": -281.43157958984375,
      "loss": 0.6931,
      "rewards/accuracies": 0.3125,
      "rewards/chosen": 7.902624929556623e-05,
      "rewards/margins": -0.0005467299488373101,
      "rewards/rejected": 0.0006257561035454273,
      "step": 10
    },
    {
      "epoch": 0.13937282229965156,
      "grad_norm": 8.21434314778786,
      "learning_rate": 4.981198836496775e-07,
      "logits/chosen": -2.7629570960998535,
      "logits/rejected": -2.7487728595733643,
      "logps/chosen": -298.9219970703125,
      "logps/rejected": -320.3416748046875,
      "loss": 0.6902,
      "rewards/accuracies": 0.581250011920929,
      "rewards/chosen": 0.009166699834167957,
      "rewards/margins": 0.006796711124479771,
      "rewards/rejected": 0.0023699889425188303,
      "step": 20
    },
    {
      "epoch": 0.20905923344947736,
      "grad_norm": 8.117010795718612,
      "learning_rate": 4.832481997086846e-07,
      "logits/chosen": -2.7146592140197754,
      "logits/rejected": -2.71540904045105,
      "logps/chosen": -296.21893310546875,
      "logps/rejected": -289.1994934082031,
      "loss": 0.6811,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": 0.04084707424044609,
      "rewards/margins": 0.0389130674302578,
      "rewards/rejected": 0.0019340096041560173,
      "step": 30
    },
    {
      "epoch": 0.2787456445993031,
      "grad_norm": 8.944785396588093,
      "learning_rate": 4.543962032878959e-07,
      "logits/chosen": -2.6810126304626465,
      "logits/rejected": -2.676293134689331,
      "logps/chosen": -298.7979736328125,
      "logps/rejected": -318.2416687011719,
      "loss": 0.6637,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.0035565136931836605,
      "rewards/margins": 0.06919760257005692,
      "rewards/rejected": -0.07275412231683731,
      "step": 40
    },
    {
      "epoch": 0.34843205574912894,
      "grad_norm": 11.10854640183049,
      "learning_rate": 4.1329321073844413e-07,
      "logits/chosen": -2.5482184886932373,
      "logits/rejected": -2.556891918182373,
      "logps/chosen": -272.1082458496094,
      "logps/rejected": -269.5940856933594,
      "loss": 0.6598,
      "rewards/accuracies": 0.59375,
      "rewards/chosen": -0.13086459040641785,
      "rewards/margins": 0.06280886381864548,
      "rewards/rejected": -0.19367346167564392,
      "step": 50
    },
    {
      "epoch": 0.34843205574912894,
      "eval_logits/chosen": -2.539210319519043,
      "eval_logits/rejected": -2.5008327960968018,
      "eval_logps/chosen": -273.58538818359375,
      "eval_logps/rejected": -286.0997619628906,
      "eval_loss": 0.6514942646026611,
      "eval_rewards/accuracies": 0.66015625,
      "eval_rewards/chosen": -0.10955476760864258,
      "eval_rewards/margins": 0.12482191622257233,
      "eval_rewards/rejected": -0.2343766838312149,
      "eval_runtime": 103.87,
      "eval_samples_per_second": 19.255,
      "eval_steps_per_second": 0.308,
      "step": 50
    },
    {
      "epoch": 0.4181184668989547,
      "grad_norm": 12.071473573808793,
      "learning_rate": 3.624028324136517e-07,
      "logits/chosen": -2.5895960330963135,
      "logits/rejected": -2.5935235023498535,
      "logps/chosen": -314.5455322265625,
      "logps/rejected": -318.70196533203125,
      "loss": 0.6509,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.11808101832866669,
      "rewards/margins": 0.14381515979766846,
      "rewards/rejected": -0.26189616322517395,
      "step": 60
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 11.064262982424012,
      "learning_rate": 3.047753100392174e-07,
      "logits/chosen": -2.533691167831421,
      "logits/rejected": -2.526494264602661,
      "logps/chosen": -305.50860595703125,
      "logps/rejected": -303.46026611328125,
      "loss": 0.6462,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.27074480056762695,
      "rewards/margins": 0.14857837557792664,
      "rewards/rejected": -0.4193231463432312,
      "step": 70
    },
    {
      "epoch": 0.5574912891986062,
      "grad_norm": 10.277351544680625,
      "learning_rate": 2.4386469286927194e-07,
      "logits/chosen": -2.562160015106201,
      "logits/rejected": -2.543243408203125,
      "logps/chosen": -298.25421142578125,
      "logps/rejected": -308.6833801269531,
      "loss": 0.6498,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.2753125727176666,
      "rewards/margins": 0.11994640529155731,
      "rewards/rejected": -0.39525893330574036,
      "step": 80
    },
    {
      "epoch": 0.627177700348432,
      "grad_norm": 10.520075906109827,
      "learning_rate": 1.8332181063127542e-07,
      "logits/chosen": -2.454864025115967,
      "logits/rejected": -2.4795737266540527,
      "logps/chosen": -267.35369873046875,
      "logps/rejected": -300.4869689941406,
      "loss": 0.6328,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.14836439490318298,
      "rewards/margins": 0.19104310870170593,
      "rewards/rejected": -0.3394075036048889,
      "step": 90
    },
    {
      "epoch": 0.6968641114982579,
      "grad_norm": 13.75424739086303,
      "learning_rate": 1.26775451942554e-07,
      "logits/chosen": -2.4828643798828125,
      "logits/rejected": -2.484675407409668,
      "logps/chosen": -328.95440673828125,
      "logps/rejected": -340.8699645996094,
      "loss": 0.6378,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.2557632327079773,
      "rewards/margins": 0.22729477286338806,
      "rewards/rejected": -0.48305803537368774,
      "step": 100
    },
    {
      "epoch": 0.6968641114982579,
      "eval_logits/chosen": -2.4382662773132324,
      "eval_logits/rejected": -2.394818067550659,
      "eval_logps/chosen": -296.12872314453125,
      "eval_logps/rejected": -319.84710693359375,
      "eval_loss": 0.6302440166473389,
      "eval_rewards/accuracies": 0.6875,
      "eval_rewards/chosen": -0.33498823642730713,
      "eval_rewards/margins": 0.2368621528148651,
      "eval_rewards/rejected": -0.5718503594398499,
      "eval_runtime": 103.4867,
      "eval_samples_per_second": 19.326,
      "eval_steps_per_second": 0.309,
      "step": 100
    },
    {
      "epoch": 0.7665505226480837,
      "grad_norm": 12.998528617340515,
      "learning_rate": 7.761486381573326e-08,
      "logits/chosen": -2.5075113773345947,
      "logits/rejected": -2.5025033950805664,
      "logps/chosen": -352.9432678222656,
      "logps/rejected": -388.5326843261719,
      "loss": 0.6258,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.41992759704589844,
      "rewards/margins": 0.22567255795001984,
      "rewards/rejected": -0.6456000208854675,
      "step": 110
    },
    {
      "epoch": 0.8362369337979094,
      "grad_norm": 11.838014849671996,
      "learning_rate": 3.878660868757322e-08,
      "logits/chosen": -2.4408838748931885,
      "logits/rejected": -2.436586856842041,
      "logps/chosen": -334.6493225097656,
      "logps/rejected": -350.96807861328125,
      "loss": 0.6344,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.4016316533088684,
      "rewards/margins": 0.2056160271167755,
      "rewards/rejected": -0.6072476506233215,
      "step": 120
    },
    {
      "epoch": 0.9059233449477352,
      "grad_norm": 10.895017244448765,
      "learning_rate": 1.261795485174083e-08,
      "logits/chosen": -2.4862332344055176,
      "logits/rejected": -2.485490322113037,
      "logps/chosen": -332.646728515625,
      "logps/rejected": -343.1838073730469,
      "loss": 0.6257,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.32124146819114685,
      "rewards/margins": 0.22806179523468018,
      "rewards/rejected": -0.5493032336235046,
      "step": 130
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 11.205634225840155,
      "learning_rate": 6.773858303274482e-10,
      "logits/chosen": -2.4655632972717285,
      "logits/rejected": -2.453521251678467,
      "logps/chosen": -299.9744567871094,
      "logps/rejected": -315.3094787597656,
      "loss": 0.6275,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": -0.3598400056362152,
      "rewards/margins": 0.13254854083061218,
      "rewards/rejected": -0.4923885762691498,
      "step": 140
    },
    {
      "epoch": 0.9965156794425087,
      "step": 143,
      "total_flos": 0.0,
      "train_loss": 0.6510118654557875,
      "train_runtime": 3809.3826,
      "train_samples_per_second": 4.814,
      "train_steps_per_second": 0.038
    }
  ],
  "logging_steps": 10,
  "max_steps": 143,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}